redict/tests/unit/moduleapi/propagate.tcl

764 lines
29 KiB
Tcl
Raw Normal View History

set testmodule [file normalize tests/modules/propagate.so]
reprocess command when client is unblocked on keys (#11012) *TL;DR* --------------------------------------- Following the discussion over the issue [#7551](https://github.com/redis/redis/issues/7551) We decided to refactor the client blocking code to eliminate some of the code duplications and to rebuild the infrastructure better for future key blocking cases. *In this PR* --------------------------------------- 1. reprocess the command once a client becomes unblocked on key (instead of running custom code for the unblocked path that's different than the one that would have run if blocking wasn't needed) 2. eliminate some (now) irrelevant code for handling unblocking lists/zsets/streams etc... 3. modify some tests to intercept the error in cases of error on reprocess after unblock (see details in the notes section below) 4. replace '$' on the client argv with current stream id. Since once we reprocess the stream XREAD we need to read from the last msg and not wait for new msg in order to prevent endless block loop. 5. Added statistics to the info "Clients" section to report the: * `total_blocking_keys` - number of blocking keys * `total_blocking_keys_on_nokey` - number of blocking keys which have at least 1 client which would like to be unblocked on when the key is deleted. 6. Avoid expiring unblocked key during unblock. Previously we used to lookup the unblocked key which might have been expired during the lookup. Now we lookup the key using NOTOUCH and NOEXPIRE to avoid deleting it at this point, so propagating commands in blocked.c is no longer needed. 7. deprecated command flags. We decided to remove the CMD_CALL_STATS and CMD_CALL_SLOWLOG and make an explicit verification in the call() function in order to decide if stats update should take place. This should simplify the logic and also mitigate existing issues: for example module calls which are triggered as part of AOF loading might still report stats even though they are called during AOF loading. *Behavior changes* --------------------------------------------------- 1. As this implementation prevents writing dedicated code handling unblocked streams/lists/zsets, since we now re-process the command once the client is unblocked some errors will be reported differently. The old implementation used to issue ``UNBLOCKED the stream key no longer exists`` in the following cases: - The stream key has been deleted (ie. calling DEL) - The stream and group existed but the key type was changed by overriding it (ie. with set command) - The key not longer exists after we swapdb with a db which does not contains this key - After swapdb when the new db has this key but with different type. In the new implementation the reported errors will be the same as if the command was processed after effect: **NOGROUP** - in case key no longer exists, or **WRONGTYPE** in case the key was overridden with a different type. 2. Reprocessing the command means that some checks will be reevaluated once the client is unblocked. For example, ACL rules might change since the command originally was executed and will fail once the client is unblocked. Another example is OOM condition checks which might enable the command to run and block but fail the command reprocess once the client is unblocked. 3. One of the changes in this PR is that no command stats are being updated once the command is blocked (all stats will be updated once the client is unblocked). This implies that when we have many clients blocked, users will no longer be able to get that information from the command stats. However the information can still be gathered from the client list. **Client blocking** --------------------------------------------------- the blocking on key will still be triggered the same way as it is done today. in order to block the current client on list of keys, the call to blockForKeys will still need to be made which will perform the same as it is today: * add the client to the list of blocked clients on each key * keep the key with a matching list node (position in the global blocking clients list for that key) in the client private blocking key dict. * flag the client with CLIENT_BLOCKED * update blocking statistics * register the client on the timeout table **Key Unblock** --------------------------------------------------- Unblocking a specific key will be triggered (same as today) by calling signalKeyAsReady. the implementation in that part will stay the same as today - adding the key to the global readyList. The reason to maintain the readyList (as apposed to iterating over all clients blocked on the specific key) is in order to keep the signal operation as short as possible, since it is called during the command processing. The main change is that instead of going through a dedicated code path that operates the blocked command we will just call processPendingCommandsAndResetClient. **ClientUnblock (keys)** --------------------------------------------------- 1. Unblocking clients on keys will be triggered after command is processed and during the beforeSleep 8. the general schema is: 9. For each key *k* in the readyList: ``` For each client *c* which is blocked on *k*: in case either: 1. *k* exists AND the *k* type matches the current client blocking type OR 2. *k* exists and *c* is blocked on module command OR 3. *k* does not exists and *c* was blocked with the flag unblock_on_deleted_key do: 1. remove the client from the list of clients blocked on this key 2. remove the blocking list node from the client blocking key dict 3. remove the client from the timeout list 10. queue the client on the unblocked_clients list 11. *NEW*: call processCommandAndResetClient(c); ``` *NOTE:* for module blocked clients we will still call the moduleUnblockClientByHandle which will queue the client for processing in moduleUnblockedClients list. **Process Unblocked clients** --------------------------------------------------- The process of all unblocked clients is done in the beforeSleep and no change is planned in that part. The general schema will be: For each client *c* in server.unblocked_clients: * remove client from the server.unblocked_clients * set back the client readHandler * continue processing the pending command and input buffer. *Some notes regarding the new implementation* --------------------------------------------------- 1. Although it was proposed, it is currently difficult to remove the read handler from the client while it is blocked. The reason is that a blocked client should be unblocked when it is disconnected, or we might consume data into void. 2. While this PR mainly keep the current blocking logic as-is, there might be some future additions to the infrastructure that we would like to have: - allow non-preemptive blocking of client - sometimes we can think that a new kind of blocking can be expected to not be preempt. for example lets imagine we hold some keys on disk and when a command needs to process them it will block until the keys are uploaded. in this case we will want the client to not disconnect or be unblocked until the process is completed (remove the client read handler, prevent client timeout, disable unblock via debug command etc...). - allow generic blocking based on command declared keys - we might want to add a hook before command processing to check if any of the declared keys require the command to block. this way it would be easier to add new kinds of key-based blocking mechanisms. Co-authored-by: Oran Agra <oran@redislabs.com> Signed-off-by: Ran Shidlansik <ranshid@amazon.com>
2023-01-01 16:35:42 -05:00
set miscmodule [file normalize tests/modules/misc.so]
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
set keyspace_events [file normalize tests/modules/keyspace_events.so]
tags "modules" {
test {Modules can propagate in async and threaded contexts} {
start_server [list overrides [list loadmodule "$testmodule"]] {
set replica [srv 0 client]
set replica_host [srv 0 host]
set replica_port [srv 0 port]
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
$replica module load $keyspace_events
start_server [list overrides [list loadmodule "$testmodule"]] {
set master [srv 0 client]
set master_host [srv 0 host]
set master_port [srv 0 port]
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
$master module load $keyspace_events
# Start the replication process...
$replica replicaof $master_host $master_port
wait_for_sync $replica
after 1000
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
test {module propagates from timer} {
set repl [attach_to_replication_stream]
$master propagate-test.timer
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
wait_for_condition 500 10 {
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
[$replica get timer] eq "3"
} else {
fail "The two counters don't match the expected value."
}
assert_replication_stream $repl {
{select *}
{incr timer}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr timer}
{incr timer}
}
close_replication_stream $repl
}
test {module propagation with notifications} {
set repl [attach_to_replication_stream]
$master set x y
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set x y}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
}
close_replication_stream $repl
}
test {module propagation with notifications with multi} {
set repl [attach_to_replication_stream]
$master multi
$master set x1 y1
$master set x2 y2
$master exec
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr notifications}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{set x1 y1}
{incr notifications}
{set x2 y2}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{exec}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
}
close_replication_stream $repl
}
test {module propagation with notifications with active-expire} {
$master debug set-active-expire 1
set repl [attach_to_replication_stream]
$master set asdf1 1 PX 300
$master set asdf2 2 PX 300
$master set asdf3 3 PX 300
wait_for_condition 500 10 {
[$replica keys asdf*] eq {}
} else {
fail "Not all keys have expired"
}
# Note whenever there's double notification: SET with PX issues two separate
# notifications: one for "set" and one for "expire"
assert_replication_stream $repl {
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr notifications}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set asdf1 1 PXAT *}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{exec}
{multi}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr notifications}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set asdf2 2 PXAT *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
{multi}
{incr notifications}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set asdf3 3 PXAT *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{multi}
{incr notifications}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr testkeyspace:expired}
{del asdf*}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
{multi}
{incr notifications}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr testkeyspace:expired}
{del asdf*}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
{multi}
{incr notifications}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr testkeyspace:expired}
{del asdf*}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
}
close_replication_stream $repl
$master debug set-active-expire 0
}
test {module propagation with notifications with eviction case 1} {
$master flushall
$master set asdf1 1
$master set asdf2 2
$master set asdf3 3
$master config set maxmemory-policy allkeys-random
$master config set maxmemory 1
# Please note the following loop:
# We evict a key and send a notification, which does INCR on the "notifications" key, so
# that every time we evict any key, "notifications" key exist (it happens inside the
# performEvictions loop). So even evicting "notifications" causes INCR on "notifications".
# If maxmemory_eviction_tenacity would have been set to 100 this would be an endless loop, but
# since the default is 10, at some point the performEvictions loop would end.
# Bottom line: "notifications" always exists and we can't really determine the order of evictions
# This test is here only for sanity
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
# The replica will get the notification with multi exec and we have a generic notification handler
# that performs `RedisModule_Call(ctx, "INCR", "c", "multi");` if the notification is inside multi exec.
# so we will have 2 keys, "notifications" and "multi".
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
wait_for_condition 500 10 {
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
[$replica dbsize] eq 2
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
} else {
fail "Not all keys have been evicted"
}
$master config set maxmemory 0
$master config set maxmemory-policy noeviction
}
test {module propagation with notifications with eviction case 2} {
$master flushall
set repl [attach_to_replication_stream]
$master set asdf1 1 EX 300
$master set asdf2 2 EX 300
$master set asdf3 3 EX 300
# Please note we use volatile eviction to prevent the loop described in the test above.
# "notifications" is not volatile so it always remains
$master config resetstat
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
$master config set maxmemory-policy volatile-ttl
$master config set maxmemory 1
wait_for_condition 500 10 {
[s evicted_keys] eq 3
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
} else {
fail "Not all keys have been evicted"
}
$master config set maxmemory 0
$master config set maxmemory-policy noeviction
$master set asdf4 4
# Note whenever there's double notification: SET with EX issues two separate
# notifications: one for "set" and one for "expire"
# Note that although CONFIG SET maxmemory is called in this flow (see issue #10014),
# eviction will happen and will not induce propagation of the CONFIG command (see #10019).
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr notifications}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set asdf1 1 PXAT *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
{multi}
{incr notifications}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set asdf2 2 PXAT *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
{multi}
{incr notifications}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set asdf3 3 PXAT *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{multi}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{del asdf*}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
{multi}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{del asdf*}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{multi}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{del asdf*}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
{multi}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set asdf4 4}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
}
close_replication_stream $repl
}
test {module propagation with timer and CONFIG SET maxmemory} {
set repl [attach_to_replication_stream]
$master config resetstat
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
$master config set maxmemory-policy volatile-random
$master propagate-test.timer-maxmemory
# Wait until the volatile keys are evicted
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
wait_for_condition 500 10 {
[s evicted_keys] eq 2
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
} else {
fail "Not all keys have been evicted"
}
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr notifications}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set timer-maxmemory-volatile-start 1 PXAT *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr timer-maxmemory-middle}
{incr notifications}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{set timer-maxmemory-volatile-end 1 PXAT *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{multi}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{del timer-maxmemory-volatile-*}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
{multi}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{del timer-maxmemory-volatile-*}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
}
close_replication_stream $repl
$master config set maxmemory 0
$master config set maxmemory-policy noeviction
}
test {module propagation with timer and EVAL} {
set repl [attach_to_replication_stream]
$master propagate-test.timer-eval
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr notifications}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incrby timer-eval-start 1}
{incr notifications}
{set foo bar}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr timer-eval-middle}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incrby timer-eval-end 1}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{exec}
}
close_replication_stream $repl
}
Fix some issues with modules and MULTI/EXEC (#8617) Bug 1: When a module ctx is freed moduleHandlePropagationAfterCommandCallback is called and handles propagation. We want to prevent it from propagating commands that were not replicated by the same context. Example: 1. module1.foo does: RM_Replicate(cmd1); RM_Call(cmd2); RM_Replicate(cmd3) 2. RM_Replicate(cmd1) propagates MULTI and adds cmd1 to also_propagagte 3. RM_Call(cmd2) create a new ctx, calls call() and destroys the ctx. 4. moduleHandlePropagationAfterCommandCallback is called, calling alsoPropagates EXEC (Note: EXEC is still not written to socket), setting server.in_trnsaction = 0 5. RM_Replicate(cmd3) is called, propagagting yet another MULTI (now we have nested MULTI calls, which is no good) and then cmd3 We must prevent RM_Call(cmd2) from resetting server.in_transaction. REDISMODULE_CTX_MULTI_EMITTED was revived for that purpose. Bug 2: Fix issues with nested RM_Call where some have '!' and some don't. Example: 1. module1.foo does RM_Call of module2.bar without replication (i.e. no '!') 2. module2.bar internally calls RM_Call of INCR with '!' 3. at the end of module1.foo we call RM_ReplicateVerbatim We want the replica/AOF to see only module1.foo and not the INCR from module2.bar Introduced a global replication_allowed flag inside RM_Call to determine whether we need to replicate or not (even if '!' was specified) Other changes: Split beforePropagateMultiOrExec to beforePropagateMulti afterPropagateExec just for better readability
2021-03-10 11:02:17 -05:00
test {module propagates nested ctx case1} {
set repl [attach_to_replication_stream]
$master propagate-test.timer-nested
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
wait_for_condition 500 10 {
Fix some issues with modules and MULTI/EXEC (#8617) Bug 1: When a module ctx is freed moduleHandlePropagationAfterCommandCallback is called and handles propagation. We want to prevent it from propagating commands that were not replicated by the same context. Example: 1. module1.foo does: RM_Replicate(cmd1); RM_Call(cmd2); RM_Replicate(cmd3) 2. RM_Replicate(cmd1) propagates MULTI and adds cmd1 to also_propagagte 3. RM_Call(cmd2) create a new ctx, calls call() and destroys the ctx. 4. moduleHandlePropagationAfterCommandCallback is called, calling alsoPropagates EXEC (Note: EXEC is still not written to socket), setting server.in_trnsaction = 0 5. RM_Replicate(cmd3) is called, propagagting yet another MULTI (now we have nested MULTI calls, which is no good) and then cmd3 We must prevent RM_Call(cmd2) from resetting server.in_transaction. REDISMODULE_CTX_MULTI_EMITTED was revived for that purpose. Bug 2: Fix issues with nested RM_Call where some have '!' and some don't. Example: 1. module1.foo does RM_Call of module2.bar without replication (i.e. no '!') 2. module2.bar internally calls RM_Call of INCR with '!' 3. at the end of module1.foo we call RM_ReplicateVerbatim We want the replica/AOF to see only module1.foo and not the INCR from module2.bar Introduced a global replication_allowed flag inside RM_Call to determine whether we need to replicate or not (even if '!' was specified) Other changes: Split beforePropagateMultiOrExec to beforePropagateMulti afterPropagateExec just for better readability
2021-03-10 11:02:17 -05:00
[$replica get timer-nested-end] eq "1"
} else {
fail "The two counters don't match the expected value."
}
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Fix some issues with modules and MULTI/EXEC (#8617) Bug 1: When a module ctx is freed moduleHandlePropagationAfterCommandCallback is called and handles propagation. We want to prevent it from propagating commands that were not replicated by the same context. Example: 1. module1.foo does: RM_Replicate(cmd1); RM_Call(cmd2); RM_Replicate(cmd3) 2. RM_Replicate(cmd1) propagates MULTI and adds cmd1 to also_propagagte 3. RM_Call(cmd2) create a new ctx, calls call() and destroys the ctx. 4. moduleHandlePropagationAfterCommandCallback is called, calling alsoPropagates EXEC (Note: EXEC is still not written to socket), setting server.in_trnsaction = 0 5. RM_Replicate(cmd3) is called, propagagting yet another MULTI (now we have nested MULTI calls, which is no good) and then cmd3 We must prevent RM_Call(cmd2) from resetting server.in_transaction. REDISMODULE_CTX_MULTI_EMITTED was revived for that purpose. Bug 2: Fix issues with nested RM_Call where some have '!' and some don't. Example: 1. module1.foo does RM_Call of module2.bar without replication (i.e. no '!') 2. module2.bar internally calls RM_Call of INCR with '!' 3. at the end of module1.foo we call RM_ReplicateVerbatim We want the replica/AOF to see only module1.foo and not the INCR from module2.bar Introduced a global replication_allowed flag inside RM_Call to determine whether we need to replicate or not (even if '!' was specified) Other changes: Split beforePropagateMultiOrExec to beforePropagateMulti afterPropagateExec just for better readability
2021-03-10 11:02:17 -05:00
{incrby timer-nested-start 1}
{incrby timer-nested-end 1}
{exec}
}
close_replication_stream $repl
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
# Note propagate-test.timer-nested just propagates INCRBY, causing an
# inconsistency, so we flush
$master flushall
Fix some issues with modules and MULTI/EXEC (#8617) Bug 1: When a module ctx is freed moduleHandlePropagationAfterCommandCallback is called and handles propagation. We want to prevent it from propagating commands that were not replicated by the same context. Example: 1. module1.foo does: RM_Replicate(cmd1); RM_Call(cmd2); RM_Replicate(cmd3) 2. RM_Replicate(cmd1) propagates MULTI and adds cmd1 to also_propagagte 3. RM_Call(cmd2) create a new ctx, calls call() and destroys the ctx. 4. moduleHandlePropagationAfterCommandCallback is called, calling alsoPropagates EXEC (Note: EXEC is still not written to socket), setting server.in_trnsaction = 0 5. RM_Replicate(cmd3) is called, propagagting yet another MULTI (now we have nested MULTI calls, which is no good) and then cmd3 We must prevent RM_Call(cmd2) from resetting server.in_transaction. REDISMODULE_CTX_MULTI_EMITTED was revived for that purpose. Bug 2: Fix issues with nested RM_Call where some have '!' and some don't. Example: 1. module1.foo does RM_Call of module2.bar without replication (i.e. no '!') 2. module2.bar internally calls RM_Call of INCR with '!' 3. at the end of module1.foo we call RM_ReplicateVerbatim We want the replica/AOF to see only module1.foo and not the INCR from module2.bar Introduced a global replication_allowed flag inside RM_Call to determine whether we need to replicate or not (even if '!' was specified) Other changes: Split beforePropagateMultiOrExec to beforePropagateMulti afterPropagateExec just for better readability
2021-03-10 11:02:17 -05:00
}
test {module propagates nested ctx case2} {
set repl [attach_to_replication_stream]
$master propagate-test.timer-nested-repl
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
wait_for_condition 500 10 {
Fix some issues with modules and MULTI/EXEC (#8617) Bug 1: When a module ctx is freed moduleHandlePropagationAfterCommandCallback is called and handles propagation. We want to prevent it from propagating commands that were not replicated by the same context. Example: 1. module1.foo does: RM_Replicate(cmd1); RM_Call(cmd2); RM_Replicate(cmd3) 2. RM_Replicate(cmd1) propagates MULTI and adds cmd1 to also_propagagte 3. RM_Call(cmd2) create a new ctx, calls call() and destroys the ctx. 4. moduleHandlePropagationAfterCommandCallback is called, calling alsoPropagates EXEC (Note: EXEC is still not written to socket), setting server.in_trnsaction = 0 5. RM_Replicate(cmd3) is called, propagagting yet another MULTI (now we have nested MULTI calls, which is no good) and then cmd3 We must prevent RM_Call(cmd2) from resetting server.in_transaction. REDISMODULE_CTX_MULTI_EMITTED was revived for that purpose. Bug 2: Fix issues with nested RM_Call where some have '!' and some don't. Example: 1. module1.foo does RM_Call of module2.bar without replication (i.e. no '!') 2. module2.bar internally calls RM_Call of INCR with '!' 3. at the end of module1.foo we call RM_ReplicateVerbatim We want the replica/AOF to see only module1.foo and not the INCR from module2.bar Introduced a global replication_allowed flag inside RM_Call to determine whether we need to replicate or not (even if '!' was specified) Other changes: Split beforePropagateMultiOrExec to beforePropagateMulti afterPropagateExec just for better readability
2021-03-10 11:02:17 -05:00
[$replica get timer-nested-end] eq "1"
} else {
fail "The two counters don't match the expected value."
}
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incrby timer-nested-start 1}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr using-call}
Fix some issues with modules and MULTI/EXEC (#8617) Bug 1: When a module ctx is freed moduleHandlePropagationAfterCommandCallback is called and handles propagation. We want to prevent it from propagating commands that were not replicated by the same context. Example: 1. module1.foo does: RM_Replicate(cmd1); RM_Call(cmd2); RM_Replicate(cmd3) 2. RM_Replicate(cmd1) propagates MULTI and adds cmd1 to also_propagagte 3. RM_Call(cmd2) create a new ctx, calls call() and destroys the ctx. 4. moduleHandlePropagationAfterCommandCallback is called, calling alsoPropagates EXEC (Note: EXEC is still not written to socket), setting server.in_trnsaction = 0 5. RM_Replicate(cmd3) is called, propagagting yet another MULTI (now we have nested MULTI calls, which is no good) and then cmd3 We must prevent RM_Call(cmd2) from resetting server.in_transaction. REDISMODULE_CTX_MULTI_EMITTED was revived for that purpose. Bug 2: Fix issues with nested RM_Call where some have '!' and some don't. Example: 1. module1.foo does RM_Call of module2.bar without replication (i.e. no '!') 2. module2.bar internally calls RM_Call of INCR with '!' 3. at the end of module1.foo we call RM_ReplicateVerbatim We want the replica/AOF to see only module1.foo and not the INCR from module2.bar Introduced a global replication_allowed flag inside RM_Call to determine whether we need to replicate or not (even if '!' was specified) Other changes: Split beforePropagateMultiOrExec to beforePropagateMulti afterPropagateExec just for better readability
2021-03-10 11:02:17 -05:00
{incr counter-1}
{incr counter-2}
{incr counter-3}
{incr counter-4}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr notifications}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr after-call}
{incr notifications}
{incr before-call-2}
{incr notifications}
{incr asdf}
{incr notifications}
{del asdf}
{incr notifications}
{incr after-call-2}
{incr notifications}
{incr timer-nested-middle}
Fix some issues with modules and MULTI/EXEC (#8617) Bug 1: When a module ctx is freed moduleHandlePropagationAfterCommandCallback is called and handles propagation. We want to prevent it from propagating commands that were not replicated by the same context. Example: 1. module1.foo does: RM_Replicate(cmd1); RM_Call(cmd2); RM_Replicate(cmd3) 2. RM_Replicate(cmd1) propagates MULTI and adds cmd1 to also_propagagte 3. RM_Call(cmd2) create a new ctx, calls call() and destroys the ctx. 4. moduleHandlePropagationAfterCommandCallback is called, calling alsoPropagates EXEC (Note: EXEC is still not written to socket), setting server.in_trnsaction = 0 5. RM_Replicate(cmd3) is called, propagagting yet another MULTI (now we have nested MULTI calls, which is no good) and then cmd3 We must prevent RM_Call(cmd2) from resetting server.in_transaction. REDISMODULE_CTX_MULTI_EMITTED was revived for that purpose. Bug 2: Fix issues with nested RM_Call where some have '!' and some don't. Example: 1. module1.foo does RM_Call of module2.bar without replication (i.e. no '!') 2. module2.bar internally calls RM_Call of INCR with '!' 3. at the end of module1.foo we call RM_ReplicateVerbatim We want the replica/AOF to see only module1.foo and not the INCR from module2.bar Introduced a global replication_allowed flag inside RM_Call to determine whether we need to replicate or not (even if '!' was specified) Other changes: Split beforePropagateMultiOrExec to beforePropagateMulti afterPropagateExec just for better readability
2021-03-10 11:02:17 -05:00
{incrby timer-nested-end 1}
{exec}
}
close_replication_stream $repl
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
# Note propagate-test.timer-nested-repl just propagates INCRBY, causing an
# inconsistency, so we flush
$master flushall
Fix some issues with modules and MULTI/EXEC (#8617) Bug 1: When a module ctx is freed moduleHandlePropagationAfterCommandCallback is called and handles propagation. We want to prevent it from propagating commands that were not replicated by the same context. Example: 1. module1.foo does: RM_Replicate(cmd1); RM_Call(cmd2); RM_Replicate(cmd3) 2. RM_Replicate(cmd1) propagates MULTI and adds cmd1 to also_propagagte 3. RM_Call(cmd2) create a new ctx, calls call() and destroys the ctx. 4. moduleHandlePropagationAfterCommandCallback is called, calling alsoPropagates EXEC (Note: EXEC is still not written to socket), setting server.in_trnsaction = 0 5. RM_Replicate(cmd3) is called, propagagting yet another MULTI (now we have nested MULTI calls, which is no good) and then cmd3 We must prevent RM_Call(cmd2) from resetting server.in_transaction. REDISMODULE_CTX_MULTI_EMITTED was revived for that purpose. Bug 2: Fix issues with nested RM_Call where some have '!' and some don't. Example: 1. module1.foo does RM_Call of module2.bar without replication (i.e. no '!') 2. module2.bar internally calls RM_Call of INCR with '!' 3. at the end of module1.foo we call RM_ReplicateVerbatim We want the replica/AOF to see only module1.foo and not the INCR from module2.bar Introduced a global replication_allowed flag inside RM_Call to determine whether we need to replicate or not (even if '!' was specified) Other changes: Split beforePropagateMultiOrExec to beforePropagateMulti afterPropagateExec just for better readability
2021-03-10 11:02:17 -05:00
}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
test {module propagates from thread} {
set repl [attach_to_replication_stream]
$master propagate-test.thread
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
wait_for_condition 500 10 {
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
[$replica get a-from-thread] eq "3"
} else {
fail "The two counters don't match the expected value."
}
assert_replication_stream $repl {
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr a-from-thread}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr thread-call}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr b-from-thread}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
{multi}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr a-from-thread}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr thread-call}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr b-from-thread}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
{multi}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr a-from-thread}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr thread-call}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr b-from-thread}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
}
close_replication_stream $repl
}
test {module propagates from thread with detached ctx} {
set repl [attach_to_replication_stream]
$master propagate-test.detached-thread
wait_for_condition 500 10 {
[$replica get thread-detached-after] eq "1"
} else {
fail "The key doesn't match the expected value."
}
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr thread-detached-before}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr notifications}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr thread-detached-1}
{incr notifications}
{incr thread-detached-2}
{incr thread-detached-after}
{exec}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
}
close_replication_stream $repl
}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
test {module propagates from command} {
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
set repl [attach_to_replication_stream]
$master propagate-test.simple
$master propagate-test.mixed
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr counter-1}
{incr counter-2}
{exec}
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr using-call}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr counter-1}
{incr counter-2}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr after-call}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
}
close_replication_stream $repl
}
test {module propagates from EVAL} {
set repl [attach_to_replication_stream]
assert_equal [ $master eval { \
redis.call("propagate-test.simple"); \
redis.call("set", "x", "y"); \
redis.call("propagate-test.mixed"); return "OK" } 0 ] {OK}
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr counter-1}
{incr counter-2}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr notifications}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{set x y}
{incr notifications}
{incr using-call}
{incr counter-1}
{incr counter-2}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr after-call}
{exec}
}
close_replication_stream $repl
}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
test {module propagates from command after good EVAL} {
set repl [attach_to_replication_stream]
assert_equal [ $master eval { return "hello" } 0 ] {hello}
$master propagate-test.simple
$master propagate-test.mixed
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
{incr counter-1}
{incr counter-2}
{exec}
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr using-call}
{incr counter-1}
{incr counter-2}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr after-call}
{exec}
}
close_replication_stream $repl
}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
test {module propagates from command after bad EVAL} {
set repl [attach_to_replication_stream]
catch { $master eval { return "hello" } -12 } e
assert_equal $e {ERR Number of keys can't be negative}
$master propagate-test.simple
$master propagate-test.mixed
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
{incr counter-1}
{incr counter-2}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{exec}
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr using-call}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr counter-1}
{incr counter-2}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr after-call}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{exec}
}
close_replication_stream $repl
}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
test {module propagates from multi-exec} {
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
set repl [attach_to_replication_stream]
$master multi
$master propagate-test.simple
$master propagate-test.mixed
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
$master propagate-test.timer-nested-repl
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
$master exec
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
wait_for_condition 500 10 {
[$replica get timer-nested-end] eq "1"
} else {
fail "The two counters don't match the expected value."
}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
assert_replication_stream $repl {
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr counter-1}
{incr counter-2}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr using-call}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr counter-1}
{incr counter-2}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr after-call}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
{multi}
{incrby timer-nested-start 1}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr using-call}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{incr counter-1}
{incr counter-2}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr counter-3}
{incr counter-4}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr notifications}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{incr after-call}
{incr notifications}
{incr before-call-2}
{incr notifications}
{incr asdf}
{incr notifications}
{del asdf}
{incr notifications}
{incr after-call-2}
{incr notifications}
{incr timer-nested-middle}
{incrby timer-nested-end 1}
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
{exec}
}
close_replication_stream $repl
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
# Note propagate-test.timer-nested just propagates INCRBY, causing an
# inconsistency, so we flush
$master flushall
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
}
test {module RM_Call of expired key propagation} {
$master debug set-active-expire 0
$master set k1 900 px 100
after 110
set repl [attach_to_replication_stream]
$master propagate-test.incr k1
assert_replication_stream $repl {
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{multi}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{select *}
{del k1}
{propagate-test.incr k1}
Sort out mess around propagation and MULTI/EXEC (#9890) The mess: Some parts use alsoPropagate for late propagation, others using an immediate one (propagate()), causing edge cases, ugly/hacky code, and the tendency for bugs The basic idea is that all commands are propagated via alsoPropagate (i.e. added to a list) and the top-most call() is responsible for going over that list and actually propagating them (and wrapping them in MULTI/EXEC if there's more than one command). This is done in the new function, propagatePendingCommands. Callers to propagatePendingCommands: 1. top-most call() (we want all nested call()s to add to the also_propagate array and just the top-most one to propagate them) - via `afterCommand` 2. handleClientsBlockedOnKeys: it is out of call() context and it may propagate stuff - via `afterCommand`. 3. handleClientsBlockedOnKeys edge case: if the looked-up key is already expired, we will propagate the expire but will not unblock any client so `afterCommand` isn't called. in that case, we have to propagate the deletion explicitly. 4. cron stuff: active-expire and eviction may also propagate stuff 5. modules: the module API allows to propagate stuff from just about anywhere (timers, keyspace notifications, threads). I could have tried to catch all the out-of-call-context places but it seemed easier to handle it in one place: when we free the context. in the spirit of what was done in call(), only the top-most freeing of a module context may cause propagation. 6. modules: when using a thread-safe ctx it's not clear when/if the ctx will be freed. we do know that the module must lock the GIL before calling RM_Replicate/RM_Call so we propagate the pending commands when releasing the GIL. A "known limitation", which were actually a bug, was fixed because of this commit (see propagate.tcl): When using a mix of RM_Call with `!` and RM_Replicate, the command would propagate out-of-order: first all the commands from RM_Call, and then the ones from RM_Replicate Another thing worth mentioning is that if, in the past, a client would issue a MULTI/EXEC with just one write command the server would blindly propagate the MULTI/EXEC too, even though it's redundant. not anymore. This commit renames propagate() to propagateNow() in order to cause conflicts in pending PRs. propagatePendingCommands is the only caller of propagateNow, which is now a static, internal helper function. Optimizations: 1. alsoPropagate will not add stuff to also_propagate if there's no AOF and replicas 2. alsoPropagate reallocs also_propagagte exponentially, to save calls to memmove Bugfixes: 1. CONFIG SET can create evictions, sending notifications which can cause to dirty++ with modules. we need to prevent it from propagating to AOF/replicas 2. We need to set current_client in RM_Call. buggy scenario: - CONFIG SET maxmemory, eviction notifications, module hook calls RM_Call - assertion in lookupKey crashes, because current_client has CONFIG SET, which isn't CMD_WRITE 3. minor: in eviction, call propagateDeletion after notification, like active-expire and all commands (we always send a notification before propagating the command)
2021-12-22 17:03:48 -05:00
{exec}
}
close_replication_stream $repl
assert_equal [$master get k1] 1
assert_equal [$master ttl k1] -1
assert_equal [$replica get k1] 1
assert_equal [$replica ttl k1] -1
}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
test {module notification on set} {
set repl [attach_to_replication_stream]
$master SADD s foo
wait_for_condition 500 10 {
[$replica SCARD s] eq "1"
} else {
fail "Failed to wait for set to be replicated"
}
$master SPOP s 1
wait_for_condition 500 10 {
[$replica SCARD s] eq "0"
} else {
fail "Failed to wait for set to be replicated"
}
# Currently the `del` command comes after the notification.
# When we fix spop to fire notification at the end (like all other commands),
# the `del` will come first.
assert_replication_stream $repl {
{multi}
{select *}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{sadd s foo}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
{multi}
{incr notifications}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{del s}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{exec}
}
close_replication_stream $repl
}
test {module key miss notification do not cause read command to be replicated} {
set repl [attach_to_replication_stream]
$master flushall
$master get unexisting_key
wait_for_condition 500 10 {
[$replica get missed] eq "1"
} else {
fail "Failed to wait for set to be replicated"
}
# Test is checking a wrong!!! behavior that causes a read command to be replicated to replica/aof.
# We keep the test to verify that such a wrong behavior does not cause any crashes.
assert_replication_stream $repl {
{select *}
{flushall}
{multi}
{incr notifications}
Reverts most of the changes of #10969 (#11178) The PR reverts the changes made on #10969. The reason for revert was trigger because of occasional test failure that started after the PR was merged. The issue is that if there is a lazy expire during the command invocation, the `del` command is added to the replication stream after the command placeholder. So the logical order on the primary is: * Delete the key (lazy expiration) * Command invocation But the replication stream gets it the other way around: * Command invocation (because the command is written into the placeholder) * Delete the key (lazy expiration) So if the command write to the key that was just lazy expired we will get inconsistency between primary and replica. One solution we considered is to add another lazy expire replication stream and write all the lazy expire there. Then when replicating, we will replicate the lazy expire replication stream first. This will solve this specific test failure but we realize that the issues does not ends here and the more we dig the more problems we find.One of the example we thought about (that can actually crashes Redis) is as follow: * User perform SINTERSTORE * When Redis tries to fetch the second input key it triggers lazy expire * The lazy expire trigger a module logic that deletes the first input key * Now Redis hold the robj of the first input key that was actually freed We believe we took the wrong approach and we will come up with another PR that solve the problem differently, for now we revert the changes so we will not have the tests failure. Notice that not the entire code was revert, some parts of the PR are changes that we would like to keep. The changes that **was** reverted are: * Saving a placeholder for replication at the beginning of the command (`call` function) * Order of the replication stream on active expire and eviction (we will decide how to handle it correctly on follow up PR) * `Spop` changes are no longer needed (because we reverted the placeholder code) Changes that **was not** reverted: * On expire/eviction, wrap the `del` and the notification effect in a multi exec. * `PropagateNow` function can still accept a special dbid, -1, indicating not to replicate select. * Keep optimisation for reusing the `alsoPropagate` array instead of allocating it each time. Tests: * All tests was kept and only few tests was modify to work correctly with the changes * Test was added to verify that the revert fixes the issues.
2022-08-24 05:51:36 -04:00
{incr missed}
Fix replication inconsistency on modules that uses key space notifications (#10969) Fix replication inconsistency on modules that uses key space notifications. ### The Problem In general, key space notifications are invoked after the command logic was executed (this is not always the case, we will discuss later about specific command that do not follow this rules). For example, the `set x 1` will trigger a `set` notification that will be invoked after the `set` logic was performed, so if the notification logic will try to fetch `x`, it will see the new data that was written. Consider the scenario on which the notification logic performs some write commands. for example, the notification logic increase some counter, `incr x{counter}`, indicating how many times `x` was changed. The logical order by which the logic was executed is has follow: ``` set x 1 incr x{counter} ``` The issue is that the `set x 1` command is added to the replication buffer at the end of the command invocation (specifically after the key space notification logic was invoked and performed the `incr` command). The replication/aof sees the commands in the wrong order: ``` incr x{counter} set x 1 ``` In this specific example the order is less important. But if, for example, the notification would have deleted `x` then we would end up with primary-replica inconsistency. ### The Solution Put the command that cause the notification in its rightful place. In the above example, the `set x 1` command logic was executed before the notification logic, so it should be added to the replication buffer before the commands that is invoked by the notification logic. To achieve this, without a major code refactoring, we save a placeholder in the replication buffer, when finishing invoking the command logic we check if the command need to be replicated, and if it does, we use the placeholder to add it to the replication buffer instead of appending it to the end. To be efficient and not allocating memory on each command to save the placeholder, the replication buffer array was modified to reuse memory (instead of allocating it each time we want to replicate commands). Also, to avoid saving a placeholder when not needed, we do it only for WRITE or MAY_REPLICATE commands. #### Additional Fixes * Expire and Eviction notifications: * Expire/Eviction logical order was to first perform the Expire/Eviction and then the notification logic. The replication buffer got this in the other way around (first notification effect and then the `del` command). The PR fixes this issue. * The notification effect and the `del` command was not wrap with `multi-exec` (if needed). The PR also fix this issue. * SPOP command: * On spop, the `spop` notification was fired before the command logic was executed. The change in this PR would have cause the replication order to be change (first `spop` command and then notification `logic`) although the logical order is first the notification logic and then the `spop` logic. The right fix would have been to move the notification to be fired after the command was executed (like all the other commands), but this can be considered a breaking change. To overcome this, the PR keeps the current behavior and changes the `spop` code to keep the right logical order when pushing commands to the replication buffer. Another PR will follow to fix the SPOP properly and match it to the other command (we split it to 2 separate PR's so it will be easy to cherry-pick this PR to 7.0 if we chose to). #### Unhanded Known Limitations * key miss event: * On key miss event, if a module performed some write command on the event (using `RM_Call`), the `dirty` counter would increase and the read command that cause the key miss event would be replicated to the replication and aof. This problem can also happened on a write command that open some keys but eventually decides not to perform any action. We decided not to handle this problem on this PR because the solution is complex and will cause additional risks in case we will want to cherry-pick this PR. We should decide if we want to handle it in future PR's. For now, modules writers is advice not to perform any write commands on key miss event. #### Testing * We already have tests to cover cases where a notification is invoking write commands that are also added to the replication buffer, the tests was modified to verify that the replica gets the command in the correct logical order. * Test was added to verify that `spop` behavior was kept unchanged. * Test was added to verify key miss event behave as expected. * Test was added to verify the changes do not break lazy expiration. #### Additional Changes * `propagateNow` function can accept a special dbid, -1, indicating not to replicate `select`. We use this to replicate `multi/exec` on `propagatePendingCommands` function. The side effect of this change is that now the `select` command will appear inside the `multi/exec` block on the replication stream (instead of outside of the `multi/exec` block). Tests was modified to match this new behavior.
2022-08-18 03:16:32 -04:00
{get unexisting_key}
{exec}
}
close_replication_stream $repl
}
test "Unload the module - propagate-test/testkeyspace" {
assert_equal {OK} [r module unload propagate-test]
assert_equal {OK} [r module unload testkeyspace]
}
assert_equal [s -1 unexpected_error_replies] 0
}
}
}
}
tags "modules aof" {
foreach aofload_type {debug_cmd startup} {
test "Modules RM_Replicate replicates MULTI/EXEC correctly: AOF-load type $aofload_type" {
start_server [list overrides [list loadmodule "$testmodule"]] {
# Enable the AOF
r config set appendonly yes
r config set auto-aof-rewrite-percentage 0 ; # Disable auto-rewrite.
waitForBgrewriteaof r
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
r propagate-test.simple
r propagate-test.mixed
r multi
Remove read-only flag from non-keyspace cmds, different approach for EXEC to propagate MULTI (#8216) In the distant history there was only the read flag for commands, and whatever command that didn't have the read flag was a write one. Then we added the write flag, but some portions of the code still used !read Also some commands that don't work on the keyspace at all, still have the read flag. Changes in this commit: 1. remove the read-only flag from TIME, ECHO, ROLE and LASTSAVE 2. EXEC command used to decides if it should propagate a MULTI by looking at the command flags (!read & !admin). When i was about to change it to look at the write flag instead, i realized that this would cause it not to propagate a MULTI for PUBLISH, EVAL, and SCRIPT, all 3 are not marked as either a read command or a write one (as they should), but all 3 are calling forceCommandPropagation. So instead of introducing a new flag to denote a command that "writes" but not into the keyspace, and still needs propagation, i decided to rely on the forceCommandPropagation, and just fix the code to propagate MULTI when needed rather than depending on the command flags at all. The implication of my change then is that now it won't decide to propagate MULTI when it sees one of these: SELECT, PING, INFO, COMMAND, TIME and other commands which are neither read nor write. 3. Changing getNodeByQuery and clusterRedirectBlockedClientIfNeeded in cluster.c to look at !write rather than read flag. This should have no implications, since these code paths are only reachable for commands which access keys, and these are always marked as either read or write. This commit improve MULTI propagation tests, for modules and a bunch of other special cases, all of which used to pass already before that commit. the only one that test change that uncovered a change of behavior is the one that DELs a non-existing key, it used to propagate an empty multi-exec block, and no longer does.
2020-12-22 05:03:49 -05:00
r propagate-test.simple
r propagate-test.mixed
r exec
assert_equal [r get counter-1] {}
assert_equal [r get counter-2] {}
assert_equal [r get using-call] 2
assert_equal [r get after-call] 2
assert_equal [r get notifications] 4
# Load the AOF
if {$aofload_type == "debug_cmd"} {
r debug loadaof
} else {
r config rewrite
restart_server 0 true false
wait_done_loading r
}
# This module behaves bad on purpose, it only calls
# RM_Replicate for counter-1 and counter-2 so values
# after AOF-load are different
assert_equal [r get counter-1] 4
assert_equal [r get counter-2] 4
assert_equal [r get using-call] 2
assert_equal [r get after-call] 2
# 4+4+2+2 commands from AOF (just above) + 4 "INCR notifications" from AOF + 4 notifications for these INCRs
assert_equal [r get notifications] 20
assert_equal {OK} [r module unload propagate-test]
assert_equal [s 0 unexpected_error_replies] 0
}
}
reprocess command when client is unblocked on keys (#11012) *TL;DR* --------------------------------------- Following the discussion over the issue [#7551](https://github.com/redis/redis/issues/7551) We decided to refactor the client blocking code to eliminate some of the code duplications and to rebuild the infrastructure better for future key blocking cases. *In this PR* --------------------------------------- 1. reprocess the command once a client becomes unblocked on key (instead of running custom code for the unblocked path that's different than the one that would have run if blocking wasn't needed) 2. eliminate some (now) irrelevant code for handling unblocking lists/zsets/streams etc... 3. modify some tests to intercept the error in cases of error on reprocess after unblock (see details in the notes section below) 4. replace '$' on the client argv with current stream id. Since once we reprocess the stream XREAD we need to read from the last msg and not wait for new msg in order to prevent endless block loop. 5. Added statistics to the info "Clients" section to report the: * `total_blocking_keys` - number of blocking keys * `total_blocking_keys_on_nokey` - number of blocking keys which have at least 1 client which would like to be unblocked on when the key is deleted. 6. Avoid expiring unblocked key during unblock. Previously we used to lookup the unblocked key which might have been expired during the lookup. Now we lookup the key using NOTOUCH and NOEXPIRE to avoid deleting it at this point, so propagating commands in blocked.c is no longer needed. 7. deprecated command flags. We decided to remove the CMD_CALL_STATS and CMD_CALL_SLOWLOG and make an explicit verification in the call() function in order to decide if stats update should take place. This should simplify the logic and also mitigate existing issues: for example module calls which are triggered as part of AOF loading might still report stats even though they are called during AOF loading. *Behavior changes* --------------------------------------------------- 1. As this implementation prevents writing dedicated code handling unblocked streams/lists/zsets, since we now re-process the command once the client is unblocked some errors will be reported differently. The old implementation used to issue ``UNBLOCKED the stream key no longer exists`` in the following cases: - The stream key has been deleted (ie. calling DEL) - The stream and group existed but the key type was changed by overriding it (ie. with set command) - The key not longer exists after we swapdb with a db which does not contains this key - After swapdb when the new db has this key but with different type. In the new implementation the reported errors will be the same as if the command was processed after effect: **NOGROUP** - in case key no longer exists, or **WRONGTYPE** in case the key was overridden with a different type. 2. Reprocessing the command means that some checks will be reevaluated once the client is unblocked. For example, ACL rules might change since the command originally was executed and will fail once the client is unblocked. Another example is OOM condition checks which might enable the command to run and block but fail the command reprocess once the client is unblocked. 3. One of the changes in this PR is that no command stats are being updated once the command is blocked (all stats will be updated once the client is unblocked). This implies that when we have many clients blocked, users will no longer be able to get that information from the command stats. However the information can still be gathered from the client list. **Client blocking** --------------------------------------------------- the blocking on key will still be triggered the same way as it is done today. in order to block the current client on list of keys, the call to blockForKeys will still need to be made which will perform the same as it is today: * add the client to the list of blocked clients on each key * keep the key with a matching list node (position in the global blocking clients list for that key) in the client private blocking key dict. * flag the client with CLIENT_BLOCKED * update blocking statistics * register the client on the timeout table **Key Unblock** --------------------------------------------------- Unblocking a specific key will be triggered (same as today) by calling signalKeyAsReady. the implementation in that part will stay the same as today - adding the key to the global readyList. The reason to maintain the readyList (as apposed to iterating over all clients blocked on the specific key) is in order to keep the signal operation as short as possible, since it is called during the command processing. The main change is that instead of going through a dedicated code path that operates the blocked command we will just call processPendingCommandsAndResetClient. **ClientUnblock (keys)** --------------------------------------------------- 1. Unblocking clients on keys will be triggered after command is processed and during the beforeSleep 8. the general schema is: 9. For each key *k* in the readyList: ``` For each client *c* which is blocked on *k*: in case either: 1. *k* exists AND the *k* type matches the current client blocking type OR 2. *k* exists and *c* is blocked on module command OR 3. *k* does not exists and *c* was blocked with the flag unblock_on_deleted_key do: 1. remove the client from the list of clients blocked on this key 2. remove the blocking list node from the client blocking key dict 3. remove the client from the timeout list 10. queue the client on the unblocked_clients list 11. *NEW*: call processCommandAndResetClient(c); ``` *NOTE:* for module blocked clients we will still call the moduleUnblockClientByHandle which will queue the client for processing in moduleUnblockedClients list. **Process Unblocked clients** --------------------------------------------------- The process of all unblocked clients is done in the beforeSleep and no change is planned in that part. The general schema will be: For each client *c* in server.unblocked_clients: * remove client from the server.unblocked_clients * set back the client readHandler * continue processing the pending command and input buffer. *Some notes regarding the new implementation* --------------------------------------------------- 1. Although it was proposed, it is currently difficult to remove the read handler from the client while it is blocked. The reason is that a blocked client should be unblocked when it is disconnected, or we might consume data into void. 2. While this PR mainly keep the current blocking logic as-is, there might be some future additions to the infrastructure that we would like to have: - allow non-preemptive blocking of client - sometimes we can think that a new kind of blocking can be expected to not be preempt. for example lets imagine we hold some keys on disk and when a command needs to process them it will block until the keys are uploaded. in this case we will want the client to not disconnect or be unblocked until the process is completed (remove the client read handler, prevent client timeout, disable unblock via debug command etc...). - allow generic blocking based on command declared keys - we might want to add a hook before command processing to check if any of the declared keys require the command to block. this way it would be easier to add new kinds of key-based blocking mechanisms. Co-authored-by: Oran Agra <oran@redislabs.com> Signed-off-by: Ran Shidlansik <ranshid@amazon.com>
2023-01-01 16:35:42 -05:00
test "Modules RM_Call does not update stats during aof load: AOF-load type $aofload_type" {
start_server [list overrides [list loadmodule "$miscmodule"]] {
# Enable the AOF
r config set appendonly yes
r config set auto-aof-rewrite-percentage 0 ; # Disable auto-rewrite.
waitForBgrewriteaof r
r config resetstat
r set foo bar
r EVAL {return redis.call('SET', KEYS[1], ARGV[1])} 1 foo bar2
r test.rm_call_replicate set foo bar3
r EVAL {return redis.call('test.rm_call_replicate',ARGV[1],KEYS[1],ARGV[2])} 1 foo set bar4
r multi
r set foo bar5
r EVAL {return redis.call('SET', KEYS[1], ARGV[1])} 1 foo bar6
r test.rm_call_replicate set foo bar7
r EVAL {return redis.call('test.rm_call_replicate',ARGV[1],KEYS[1],ARGV[2])} 1 foo set bar8
r exec
assert_match {*calls=8,*,rejected_calls=0,failed_calls=0} [cmdrstat set r]
# Load the AOF
if {$aofload_type == "debug_cmd"} {
r config resetstat
r debug loadaof
} else {
r config rewrite
restart_server 0 true false
wait_done_loading r
}
assert_no_match {*calls=*} [cmdrstat set r]
}
}
}
}