For diskless replication in swapdb mode, considering we already spend replica memory
having a backup of current db to restore in case of failure, we can have the following benefits
by instead swapping database only in case we succeeded in transferring db from master:
- Avoid `LOADING` response during failed and successful synchronization for cases where the
replica is already up and running with data.
- Faster total time of diskless replication, because now we're moving from Transfer + Flush + Load
time to Transfer + Load only. Flushing the tempDb is done asynchronously after swapping.
- This could be implemented also for disk replication with similar benefits if consumers are willing
to spend the extra memory usage.
General notes:
- The concept of `backupDb` becomes `tempDb` for clarity.
- Async loading mode will only kick in if the replica is syncing from a master that has the same
repl-id the one it had before. i.e. the data it's getting belongs to a different time of the same timeline.
- New property in INFO: `async_loading` to differentiate from the blocking loading
- Slot to Key mapping is now a field of `redisDb` as it's more natural to access it from both server.db
and the tempDb that is passed around.
- Because this is affecting replicas only, we assume that if they are not readonly and write commands
during replication, they are lost after SYNC same way as before, but we're still denying CONFIG SET
here anyways to avoid complications.
Considerations for review:
- We have many cases where server.loading flag is used and even though I tried my best, there may
be cases where async_loading should be checked as well and cases where it shouldn't (would require
very good understanding of whole code)
- Several places that had different behavior depending on the loading flag where actually meant to just
handle commands coming from the AOF client differently than ones coming from real clients, changed
to check CLIENT_ID_AOF instead.
**Additional for Release Notes**
- Bugfix - server.dirty was not incremented for any kind of diskless replication, as effect it wouldn't
contribute on triggering next database SAVE
- New flag for RM_GetContextFlags module API: REDISMODULE_CTX_FLAGS_ASYNC_LOADING
- Deprecated RedisModuleEvent_ReplBackup. Starting from Redis 7.0, we don't fire this event.
Instead, we have the new RedisModuleEvent_ReplAsyncLoad holding 3 sub-events: STARTED,
ABORTED and COMPLETED.
- New module flag REDISMODULE_OPTIONS_HANDLE_REPL_ASYNC_LOAD for RedisModule_SetModuleOptions
to allow modules to declare they support the diskless replication with async loading (when absent, we fall
back to disk-based loading).
Co-authored-by: Eduardo Semprebon <edus@saxobank.com>
Co-authored-by: Oran Agra <oran@redislabs.com>
Add new no-mandatory-keys flag to support COMMAND GETKEYS of commands
which have no mandatory keys.
In the past we would have got this error:
```
127.0.0.1:6379> command getkeys eval "return 1" 0
(error) ERR Invalid arguments specified for command
```
When using SETNX and SETXX we could end up doing key lookup twice.
This presents a small inefficiency price.
Also once we have statistics of write hit and miss they'll be wrong (recording the same key hit twice)
After PR #9166 , replication backlog is not a real block of memory, just contains a
reference points to replication buffer's block and the blocks index (to accelerate
search offset when partial sync), so we need update both replication buffer's block's
offset and replication backlog blocks index's offset when master restart from RDB,
since the `server.master_repl_offset` is changed.
The implications of this bug was just a slow search, but not a replication failure.
Add timestamp annotation in AOF, one part of #9325.
Enabled with the new `aof-timestamp-enabled` config option.
Timestamp annotation format is "#TS:${timestamp}\r\n"."
TS" is short of timestamp and this method could save extra bytes in AOF.
We can use timestamp annotation for some special functions.
- know the executing time of commands
- restore data to a specific point-in-time (by using redis-check-rdb to truncate the file)
Improve code doc for allowed_firstargs (used to be allowed_commands before #9504.
I don't think the text in the code needs to refer to the history (it's not there just for backwards compatibility).
instead it should just describe what it does.
## Background
For redis master, one replica uses one copy of replication buffer, that is a big waste of memory,
more replicas more waste, and allocate/free memory for every reply list also cost much.
If we set client-output-buffer-limit small and write traffic is heavy, master may disconnect with
replicas and can't finish synchronization with replica. If we set client-output-buffer-limit big,
master may be OOM when there are many replicas that separately keep much memory.
Because replication buffers of different replica client are the same, one simple idea is that
all replicas only use one replication buffer, that will effectively save memory.
Since replication backlog content is the same as replicas' output buffer, now we
can discard replication backlog memory and use global shared replication buffer
to implement replication backlog mechanism.
## Implementation
I create one global "replication buffer" which contains content of replication stream.
The structure of "replication buffer" is similar to the reply list that exists in every client.
But the node of list is `replBufBlock`, which has `id, repl_offset, refcount` fields.
```c
/* Replication buffer blocks is the list of replBufBlock.
*
* +--------------+ +--------------+ +--------------+
* | refcount = 1 | ... | refcount = 0 | ... | refcount = 2 |
* +--------------+ +--------------+ +--------------+
* | / \
* | / \
* | / \
* Repl Backlog Replia_A Replia_B
*
* Each replica or replication backlog increments only the refcount of the
* 'ref_repl_buf_node' which it points to. So when replica walks to the next
* node, it should first increase the next node's refcount, and when we trim
* the replication buffer nodes, we remove node always from the head node which
* refcount is 0. If the refcount of the head node is not 0, we must stop
* trimming and never iterate the next node. */
/* Similar with 'clientReplyBlock', it is used for shared buffers between
* all replica clients and replication backlog. */
typedef struct replBufBlock {
int refcount; /* Number of replicas or repl backlog using. */
long long id; /* The unique incremental number. */
long long repl_offset; /* Start replication offset of the block. */
size_t size, used;
char buf[];
} replBufBlock;
```
So now when we feed replication stream into replication backlog and all replicas, we only need
to feed stream into replication buffer `feedReplicationBuffer`. In this function, we set some fields of
replication backlog and replicas to references of the global replication buffer blocks. And we also
need to check replicas' output buffer limit to free if exceeding `client-output-buffer-limit`, and trim
replication backlog if exceeding `repl-backlog-size`.
When sending reply to replicas, we also need to iterate replication buffer blocks and send its
content, when totally sending one block for replica, we decrease current node count and
increase the next current node count, and then free the block which reference is 0 from the
head of replication buffer blocks.
Since now we use linked list to manage replication backlog, it may cost much time for iterating
all linked list nodes to find corresponding replication buffer node. So we create a rax tree to
store some nodes for index, but to avoid rax tree occupying too much memory, i record
one per 64 nodes for index.
Currently, to make partial resynchronization as possible as much, we always let replication
backlog as the last reference of replication buffer blocks, backlog size may exceeds our setting
if slow replicas that reference vast replication buffer blocks, and this method doesn't increase
memory usage since they share replication buffer. To avoid freezing server for freeing unreferenced
replication buffer blocks when we need to trim backlog for exceeding backlog size setting,
we trim backlog incrementally (free 64 blocks per call now), and make it faster in
`beforeSleep` (free 640 blocks).
### Other changes
- `mem_total_replication_buffers`: we add this field in INFO command, it means the total
memory of replication buffers used.
- `mem_clients_slaves`: now even replica is slow to replicate, and its output buffer memory
is not 0, but it still may be 0, since replication backlog and replicas share one global replication
buffer, only if replication buffer memory is more than the repl backlog setting size, we consider
the excess as replicas' memory. Otherwise, we think replication buffer memory is the consumption
of repl backlog.
- Key eviction
Since all replicas and replication backlog share global replication buffer, we think only the
part of exceeding backlog size the extra separate consumption of replicas.
Because we trim backlog incrementally in the background, backlog size may exceeds our
setting if slow replicas that reference vast replication buffer blocks disconnect.
To avoid massive eviction loop, we don't count the delayed freed replication backlog into
used memory even if there are no replicas, i.e. we also regard this memory as replicas's memory.
- `client-output-buffer-limit` check for replica clients
It doesn't make sense to set the replica clients output buffer limit lower than the repl-backlog-size
config (partial sync will succeed and then replica will get disconnected). Such a configuration is
ignored (the size of repl-backlog-size will be used). This doesn't have memory consumption
implications since the replica client will share the backlog buffers memory.
- Drop replication backlog after loading data if needed
We always create replication backlog if server is a master, we need it because we put DELs in
it when loading expired keys in RDB, but if RDB doesn't have replication info or there is no rdb,
it is not possible to support partial resynchronization, to avoid extra memory of replication backlog,
we drop it.
- Multi IO threads
Since all replicas and replication backlog use global replication buffer, if I/O threads are enabled,
to guarantee data accessing thread safe, we must let main thread handle sending the output buffer
to all replicas. But before, other IO threads could handle sending output buffer of all replicas.
## Other optimizations
This solution resolve some other problem:
- When replicas disconnect with master since of out of output buffer limit, releasing the output
buffer of replicas may freeze server if we set big `client-output-buffer-limit` for replicas, but now,
it doesn't cause freezing.
- This implementation may mitigate reply list copy cost time(also freezes server) when one replication
has huge reply buffer and another replica can copy buffer for full synchronization. now, we just copy
reference info, it is very light.
- If we set replication backlog size big, it also may cost much time to copy replication backlog into
replica's output buffer. But this commit eliminates this problem.
- Resizing replication backlog size doesn't empty current replication backlog content.
## Intro
The purpose is to allow having different flags/ACL categories for
subcommands (Example: CONFIG GET is ok-loading but CONFIG SET isn't)
We create a small command table for every command that has subcommands
and each subcommand has its own flags, etc. (same as a "regular" command)
This commit also unites the Redis and the Sentinel command tables
## Affected commands
CONFIG
Used to have "admin ok-loading ok-stale no-script"
Changes:
1. Dropped "ok-loading" in all except GET (this doesn't change behavior since
there were checks in the code doing that)
XINFO
Used to have "read-only random"
Changes:
1. Dropped "random" in all except CONSUMERS
XGROUP
Used to have "write use-memory"
Changes:
1. Dropped "use-memory" in all except CREATE and CREATECONSUMER
COMMAND
No changes.
MEMORY
Used to have "random read-only"
Changes:
1. Dropped "random" in PURGE and USAGE
ACL
Used to have "admin no-script ok-loading ok-stale"
Changes:
1. Dropped "admin" in WHOAMI, GENPASS, and CAT
LATENCY
No changes.
MODULE
No changes.
SLOWLOG
Used to have "admin random ok-loading ok-stale"
Changes:
1. Dropped "random" in RESET
OBJECT
Used to have "read-only random"
Changes:
1. Dropped "random" in ENCODING and REFCOUNT
SCRIPT
Used to have "may-replicate no-script"
Changes:
1. Dropped "may-replicate" in all except FLUSH and LOAD
CLIENT
Used to have "admin no-script random ok-loading ok-stale"
Changes:
1. Dropped "random" in all except INFO and LIST
2. Dropped "admin" in ID, TRACKING, CACHING, GETREDIR, INFO, SETNAME, GETNAME, and REPLY
STRALGO
No changes.
PUBSUB
No changes.
CLUSTER
Changes:
1. Dropped "admin in countkeysinslots, getkeysinslot, info, nodes, keyslot, myid, and slots
SENTINEL
No changes.
(note that DEBUG also fits, but we decided not to convert it since it's for
debugging and anyway undocumented)
## New sub-command
This commit adds another element to the per-command output of COMMAND,
describing the list of subcommands, if any (in the same structure as "regular" commands)
Also, it adds a new subcommand:
```
COMMAND LIST [FILTERBY (MODULE <module-name>|ACLCAT <cat>|PATTERN <pattern>)]
```
which returns a set of all commands (unless filters), but excluding subcommands.
## Module API
A new module API, RM_CreateSubcommand, was added, in order to allow
module writer to define subcommands
## ACL changes:
1. Now, that each subcommand is actually a command, each has its own ACL id.
2. The old mechanism of allowed_subcommands is redundant
(blocking/allowing a subcommand is the same as blocking/allowing a regular command),
but we had to keep it, to support the widespread usage of allowed_subcommands
to block commands with certain args, that aren't subcommands (e.g. "-select +select|0").
3. I have renamed allowed_subcommands to allowed_firstargs to emphasize the difference.
4. Because subcommands are commands in ACL too, you can now use "-" to block subcommands
(e.g. "+client -client|kill"), which wasn't possible in the past.
5. It is also possible to use the allowed_firstargs mechanism with subcommand.
For example: `+config -config|set +config|set|loglevel` will block all CONFIG SET except
for setting the log level.
6. All of the ACL changes above required some amount of refactoring.
## Misc
1. There are two approaches: Either each subcommand has its own function or all
subcommands use the same function, determining what to do according to argv[0].
For now, I took the former approaches only with CONFIG and COMMAND,
while other commands use the latter approach (for smaller blamelog diff).
2. Deleted memoryGetKeys: It is no longer needed because MEMORY USAGE now uses the "range" key spec.
4. Bugfix: GETNAME was missing from CLIENT's help message.
5. Sentinel and Redis now use the same table, with the same function pointer.
Some commands have a different implementation in Sentinel, so we redirect
them (these are ROLE, PUBLISH, and INFO).
6. Command stats now show the stats per subcommand (e.g. instead of stats just
for "config" you will have stats for "config|set", "config|get", etc.)
7. It is now possible to use COMMAND directly on subcommands:
COMMAND INFO CONFIG|GET (The pipeline syntax was inspired from ACL, and
can be used in functions lookupCommandBySds and lookupCommandByCString)
8. STRALGO is now a container command (has "help")
## Breaking changes:
1. Command stats now show the stats per subcommand (see (5) above)
Since the size of mode_t is platform dependant we handle the
`unixsocketperm` configuration as a generic int type.
mode_t is either an unsigned int or unsigned short (macOS) and
the range-limits allows for a simple cast to a mode_t.
This is useful for approximating size computation of complex module types.
Note that the mem_usage2 callback is new and has not been released yet, which is why we can modify it.
Tracking invalidation messages were sometimes sent in inconsistent order,
before the command's reply rather than after.
In addition to that, they were sometimes embedded inside other commands
responses, like MULTI-EXEC and MGET.
When queuing a multi command we duplicated the argv (meaning an alloc
and a memcpy). This isn't needed since we can use the previously allocated
argv and just reset the client objects argv to NULL. This should saves some
memory and is a minor optimization in heavy MULTI/EXEC traffic, especially
if there are lots of arguments.
- fix possible heap corruption in ziplist and listpack resulting by trying to
allocate more than the maximum size of 4GB.
- prevent ziplist (hash and zset) from reaching size of above 1GB, will be
converted to HT encoding, that's not a useful size.
- prevent listpack (stream) from reaching size of above 1GB.
- XADD will start a new listpack if the new record may cause the previous
listpack to grow over 1GB.
- XADD will respond with an error if a single stream record is over 1GB
- List type (ziplist in quicklist) was truncating strings that were over 4GB,
now it'll respond with an error.
Co-authored-by: sundb <sundbcn@gmail.com>
This change sets a low limit for multibulk and bulk length in the
protocol for unauthenticated connections, so that they can't easily
cause redis to allocate massive amounts of memory by sending just a few
characters on the network.
The new limits are 10 arguments of 16kb each (instead of 1m of 512mb)
Remove hard coded multi-bulk limit (was 1,048,576), new limit is INT_MAX.
When client sends an m-bulk that's higher than 1024, we initially only allocate
the argv array for 1024 arguments, and gradually grow that allocation as arguments
are received.
1. Remove forward declarations from header files to functions that do not exist:
hmsetCommand and rdbSaveTime.
2. Minor phrasing fixes in #9519
3. Add missing sdsfree(title) and fix typo in redis-benchmark.
4. Modify some error comments in some zset commands.
5. Fix copy-paste bug comment in syncWithMaster about `ip-address`.
Fixing CI test issues introduced in #8687
- valgrind warnings in readQueryFromClient when client was freed by processInputBuffer
- adding DEBUG pause-cron for tests not to be time dependent.
- skipping a test that depends on socket buffers / events not compatible with TLS
- making sure client got subscribed by not using deferring client
### Description
A mechanism for disconnecting clients when the sum of all connected clients is above a
configured limit. This prevents eviction or OOM caused by accumulated used memory
between all clients. It's a complimentary mechanism to the `client-output-buffer-limit`
mechanism which takes into account not only a single client and not only output buffers
but rather all memory used by all clients.
#### Design
The general design is as following:
* We track memory usage of each client, taking into account all memory used by the
client (query buffer, output buffer, parsed arguments, etc...). This is kept up to date
after reading from the socket, after processing commands and after writing to the socket.
* Based on the used memory we sort all clients into buckets. Each bucket contains all
clients using up up to x2 memory of the clients in the bucket below it. For example up
to 1m clients, up to 2m clients, up to 4m clients, ...
* Before processing a command and before sleep we check if we're over the configured
limit. If we are we start disconnecting clients from larger buckets downwards until we're
under the limit.
#### Config
`maxmemory-clients` max memory all clients are allowed to consume, above this threshold
we disconnect clients.
This config can either be set to 0 (meaning no limit), a size in bytes (possibly with MB/GB
suffix), or as a percentage of `maxmemory` by using the `%` suffix (e.g. setting it to `10%`
would mean 10% of `maxmemory`).
#### Important code changes
* During the development I encountered yet more situations where our io-threads access
global vars. And needed to fix them. I also had to handle keeps the clients sorted into the
memory buckets (which are global) while their memory usage changes in the io-thread.
To achieve this I decided to simplify how we check if we're in an io-thread and make it
much more explicit. I removed the `CLIENT_PENDING_READ` flag used for checking
if the client is in an io-thread (it wasn't used for anything else) and just used the global
`io_threads_op` variable the same way to check during writes.
* I optimized the cleanup of the client from the `clients_pending_read` list on client freeing.
We now store a pointer in the `client` struct to this list so we don't need to search in it
(`pending_read_list_node`).
* Added `evicted_clients` stat to `INFO` command.
* Added `CLIENT NO-EVICT ON|OFF` sub command to exclude a specific client from the
client eviction mechanism. Added corrosponding 'e' flag in the client info string.
* Added `multi-mem` field in the client info string to show how much memory is used up
by buffered multi commands.
* Client `tot-mem` now accounts for buffered multi-commands, pubsub patterns and
channels (partially), tracking prefixes (partially).
* CLIENT_CLOSE_ASAP flag is now handled in a new `beforeNextClient()` function so
clients will be disconnected between processing different clients and not only before sleep.
This new function can be used in the future for work we want to do outside the command
processing loop but don't want to wait for all clients to be processed before we get to it.
Specifically I wanted to handle output-buffer-limit related closing before we process client
eviction in case the two race with each other.
* Added a `DEBUG CLIENT-EVICTION` command to print out info about the client eviction
buckets.
* Each client now holds a pointer to the client eviction memory usage bucket it belongs to
and listNode to itself in that bucket for quick removal.
* Global `io_threads_op` variable now can contain a `IO_THREADS_OP_IDLE` value
indicating no io-threading is currently being executed.
* In order to track memory used by each clients in real-time we can't rely on updating
these stats in `clientsCron()` alone anymore. So now I call `updateClientMemUsage()`
(used to be `clientsCronTrackClientsMemUsage()`) after command processing, after
writing data to pubsub clients, after writing the output buffer and after reading from the
socket (and maybe other places too). The function is written to be fast.
* Clients are evicted if needed (with appropriate log line) in `beforeSleep()` and before
processing a command (before performing oom-checks and key-eviction).
* All clients memory usage buckets are grouped as follows:
* All clients using less than 64k.
* 64K..128K
* 128K..256K
* ...
* 2G..4G
* All clients using 4g and up.
* Added client-eviction.tcl with a bunch of tests for the new mechanism.
* Extended maxmemory.tcl to test the interaction between maxmemory and
maxmemory-clients settings.
* Added an option to flag a numeric configuration variable as a "percent", this means that
if we encounter a '%' after the number in the config file (or config set command) we
consider it as valid. Such a number is store internally as a negative value. This way an
integer value can be interpreted as either a percent (negative) or absolute value (positive).
This is useful for example if some numeric configuration can optionally be set to a percentage
of something else.
Co-authored-by: Oran Agra <oran@redislabs.com>
This commit introduced a new flag to the RM_Call:
'C' - Check if the command can be executed according to the ACLs associated with it.
Also, three new API's added to check if a command, key, or channel can be executed or accessed
by a user, according to the ACLs associated with it.
- RM_ACLCheckCommandPerm
- RM_ACLCheckKeyPerm
- RM_ACLCheckChannelPerm
The user for these API's is a RedisModuleUser object, that for a Module user returned by the RM_CreateModuleUser API, or for a general ACL user can be retrieved by these two new API's:
- RM_GetCurrentUserName - Retrieve the user name of the client connection behind the current context.
- RM_GetModuleUserFromUserName - Get a RedisModuleUser from a user name
As a result of getting a RedisModuleUser from name, it can now also access the general ACL users (not just ones created by the module).
This mean the already existing API RM_SetModuleUserACL(), can be used to change the ACL rules for such users.
This is similar to the recent addition of LMPOP/BLMPOP (#9373), but zset.
Syntax for the new ZMPOP command:
`ZMPOP numkeys [<key> ...] MIN|MAX [COUNT count]`
Syntax for the new BZMPOP command:
`BZMPOP timeout numkeys [<key> ...] MIN|MAX [COUNT count]`
Some background:
- ZPOPMIN/ZPOPMAX take only one key, and can return multiple elements.
- BZPOPMIN/BZPOPMAX take multiple keys, but return only one element from just one key.
- ZMPOP/BZMPOP can take multiple keys, and can return multiple elements from just one key.
Note that ZMPOP/BZMPOP can take multiple keys, it eventually operates on just on key.
And it will propagate as ZPOPMIN or ZPOPMAX with the COUNT option.
As new commands, if we can not pop any elements, the response like:
- ZMPOP: Return a NIL in both RESP2 and RESP3, unlike ZPOPMIN/ZPOPMAX return emptyarray.
- BZMPOP: Return a NIL in both RESP2 and RESP3 when timeout is reached, like BZPOPMIN/BZPOPMAX.
For the normal response is nested arrays in RESP2 and RESP3:
```
ZMPOP/BZMPOP
1) keyname
2) 1) 1) member1
2) score1
2) 1) member2
2) score2
In RESP2:
1) "myzset"
2) 1) 1) "three"
2) "3"
2) 1) "two"
2) "2"
In RESP3:
1) "myzset"
2) 1) 1) "three"
2) (double) 3
2) 1) "two"
2) (double) 2
```
Implements the [LIMIT limit] variant of SINTERCARD/ZINTERCARD.
Now with the LIMIT, we can stop the searching when cardinality
reaching the limit, and return the cardinality ASAP.
Note that in SINTERCARD, the old synatx was: `SINTERCARD key [key ...]`
In order to add a optional parameter, we must break the old synatx.
So the new syntax of SINTERCARD will be consistent with ZINTERCARD.
New syntax: `SINTERCARD numkeys key [key ...] [LIMIT limit]`.
Note that this means that SINTERCARD has a different syntax than
SINTER and SINTERSTORE (taking numkeys argument)
As for ZINTERCARD, we can easily add a optional parameter to it.
New syntax: `ZINTERCARD numkeys key [key ...] [LIMIT limit]`
The `cmd` argument was completely unused, and all the code that bothered to pass it was unnecessary.
This is a prepartion for a future commit that treats subcommands as commands
Fix#7297
The problem:
Today, there is no way for a client library or app to know the key name indexes for commands such as
ZUNIONSTORE/EVAL and others with "numkeys", since COMMAND INFO returns no useful info for them.
For cluster-aware redis clients, this requires to 'patch' the client library code specifically for each of these commands or to
resolve each execution of these commands with COMMAND GETKEYS.
The solution:
Introducing key specs other than the legacy "range" (first,last,step)
The 8th element of the command info array, if exists, holds an array of key specs. The array may be empty, which indicates
the command doesn't take any key arguments or may contain one or more key-specs, each one may leads to the discovery
of 0 or more key arguments.
A client library that doesn't support this key-spec feature will keep using the first,last,step and movablekeys flag which will
obviously remain unchanged.
A client that supports this key-specs feature needs only to look at the key-specs array. If it finds an unrecognized spec, it
must resort to using COMMAND GETKEYS if it wishes to get all key name arguments, but if all it needs is one key in order
to know which cluster node to use, then maybe another spec (if the command has several) can supply that, and there's no
need to use GETKEYS.
Each spec is an array of arguments, first one is the spec name, the second is an array of flags, and the third is an array
containing details about the spec (specific meaning for each spec type)
The initial flags we support are "read" and "write" indicating if the keys that this key-spec finds are used for read or for write.
clients should ignore any unfamiliar flags.
In order to easily find the positions of keys in a given array of args we introduce keys specs. There are two logical steps of
key specs:
1. `start_search`: Given an array of args, indicate where we should start searching for keys
2. `find_keys`: Given the output of start_search and an array of args, indicate all possible indices of keys.
### start_search step specs
- `index`: specify an argument index explicitly
- `index`: 0 based index (1 means the first command argument)
- `keyword`: specify a string to match in `argv`. We should start searching for keys just after the keyword appears.
- `keyword`: the string to search for
- `start_search`: an index from which to start the keyword search (can be negative, which means to search from the end)
Examples:
- `SET` has start_search of type `index` with value `1`
- `XREAD` has start_search of type `keyword` with value `[“STREAMS”,1]`
- `MIGRATE` has start_search of type `keyword` with value `[“KEYS”,-2]`
### find_keys step specs
- `range`: specify `[count, step, limit]`.
- `lastkey`: index of the last key. relative to the index returned from begin_search. -1 indicating till the last argument, -2 one before the last
- `step`: how many args should we skip after finding a key, in order to find the next one
- `limit`: if count is -1, we use limit to stop the search by a factor. 0 and 1 mean no limit. 2 means ½ of the remaining args, 3 means ⅓, and so on.
- “keynum”: specify `[keynum_index, first_key_index, step]`.
- `keynum_index`: is relative to the return of the `start_search` spec.
- `first_key_index`: is relative to `keynum_index`.
- `step`: how many args should we skip after finding a key, in order to find the next one
Examples:
- `SET` has `range` of `[0,1,0]`
- `MSET` has `range` of `[-1,2,0]`
- `XREAD` has `range` of `[-1,1,2]`
- `ZUNION` has `start_search` of type `index` with value `1` and `find_keys` of type `keynum` with value `[0,1,1]`
- `AI.DAGRUN` has `start_search` of type `keyword` with value `[“LOAD“,1]` and `find_keys` of type `keynum` with value
`[0,1,1]` (see https://oss.redislabs.com/redisai/master/commands/#aidagrun)
Note: this solution is not perfect as the module writers can come up with anything, but at least we will be able to find the key
args of the vast majority of commands.
If one of the above specs can’t describe the key positions, the module writer can always fall back to the `getkeys-api` option.
Some keys cannot be found easily (`KEYS` in `MIGRATE`: Imagine the argument for `AUTH` is the string “KEYS” - we will
start searching in the wrong index).
The guarantee is that the specs may be incomplete (`incomplete` will be specified in the spec to denote that) but we never
report false information (assuming the command syntax is correct).
For `MIGRATE` we start searching from the end - `startfrom=-1` - and if one of the keys is actually called "keys" we will
report only a subset of all keys - hence the `incomplete` flag.
Some `incomplete` specs can be completely empty (i.e. UNKNOWN begin_search) which should tell the client that
COMMAND GETKEYS (or any other way to get the keys) must be used (Example: For `SORT` there is no way to describe
the STORE keyword spec, as the word "store" can appear anywhere in the command).
We will expose these key specs in the `COMMAND` command so that clients can learn, on startup, where the keys are for
all commands instead of holding hardcoded tables or use `COMMAND GETKEYS` in runtime.
Comments:
1. Redis doesn't internally use the new specs, they are only used for COMMAND output.
2. In order to support the current COMMAND INFO format (reply array indices 4, 5, 6) we created a synthetic range, called
legacy_range, that, if possible, is built according to the new specs.
3. Redis currently uses only getkeys_proc or the legacy_range to get the keys indices (in COMMAND GETKEYS for
example).
"incomplete" specs:
the command we have issues with are MIGRATE, STRALGO, and SORT
for MIGRATE, because the token KEYS, if exists, must be the last token, we can search in reverse. it one of the keys is
actually the string "keys" will return just a subset of the keys (hence, it's "incomplete")
for SORT and STRALGO we can use this heuristic (the keys can be anywhere in the command) and therefore we added a
key spec that is both "incomplete" and of "unknown type"
if a client encounters an "incomplete" spec it means that it must find a different way (either COMMAND GETKEYS or have
its own parser) to retrieve the keys.
please note that all commands, apart from the three mentioned above, have "complete" key specs
List functions operating on elements by index:
* RM_ListGet
* RM_ListSet
* RM_ListInsert
* RM_ListDelete
Iteration is done using a simple for loop over indices.
The index based functions use an internal iterator as an optimization.
This is explained in the docs:
```
* Many of the list functions access elements by index. Since a list is in
* essence a doubly-linked list, accessing elements by index is generally an
* O(N) operation. However, if elements are accessed sequentially or with
* indices close together, the functions are optimized to seek the index from
* the previous index, rather than seeking from the ends of the list.
*
* This enables iteration to be done efficiently using a simple for loop:
*
* long n = RM_ValueLength(key);
* for (long i = 0; i < n; i++) {
* RedisModuleString *elem = RedisModule_ListGet(key, i);
* // Do stuff...
* }
```
The main idea is how to allow a master to load replication info from RDB file when rebooting, if master can load replication info it means that replicas may have the chance to psync with master, it can save much traffic.
The key point is we need guarantee safety and consistency, so there
are two differences between master and replica:
1. master would load the replication info as secondary ID and
offset, in case other masters have the same replid.
2. when master loading RDB, it would propagate expired keys as DEL
command to replication backlog, then replica can receive these
commands to delete stale keys.
p.s. the expired keys when RDB loading is useful for users, so
we show it as `rdb_last_load_keys_expired` and `rdb_last_load_keys_loaded` in info persistence.
Moreover, after load replication info, master should update
`no_replica_time` in case loading RDB cost too long time.
Part two of implementing #8702 (zset), after #8887.
## Description of the feature
Replaced all uses of ziplist with listpack in t_zset, and optimized some of the code to optimize performance.
## Rdb format changes
New `RDB_TYPE_ZSET_LISTPACK` rdb type.
## Rdb loading improvements:
1) Pre-expansion of dict for validation of duplicate data for listpack and ziplist.
2) Simplifying the release of empty key objects when RDB loading.
3) Unify ziplist and listpack data verify methods for zset and hash, and move code to rdb.c.
## Interface changes
1) New `zset-max-listpack-entries` config is an alias for `zset-max-ziplist-entries` (same with `zset-max-listpack-value`).
2) OBJECT ENCODING will return listpack instead of ziplist.
## Listpack improvements:
1) Add `lpDeleteRange` and `lpDeleteRangeWithEntry` functions to delete a range of entries from listpack.
2) Improve the performance of `lpCompare`, converting from string to integer is faster than converting from integer to string.
3) Replace `snprintf` with `ll2string` to improve performance in converting numbers to strings in `lpGet()`.
## Zset improvements:
1) Improve the performance of `zzlFind` method, use `lpFind` instead of `lpCompare` in a loop.
2) Use `lpDeleteRangeWithEntry` instead of `lpDelete` twice to delete a element of zset.
## Tests
1) Add some unittests for `lpDeleteRange` and `lpDeleteRangeWithEntry` function.
2) Add zset RDB loading test.
3) Add benchmark test for `lpCompare` and `ziplsitCompare`.
4) Add empty listpack zset corrupt dump test.
We want to add COUNT option for BLPOP.
But we can't do it without breaking compatibility due to the command arguments syntax.
So this commit introduce two new commands.
Syntax for the new LMPOP command:
`LMPOP numkeys [<key> ...] LEFT|RIGHT [COUNT count]`
Syntax for the new BLMPOP command:
`BLMPOP timeout numkeys [<key> ...] LEFT|RIGHT [COUNT count]`
Some background:
- LPOP takes one key, and can return multiple elements.
- BLPOP takes multiple keys, but returns one element from just one key.
- LMPOP can take multiple keys and return multiple elements from just one key.
Note that LMPOP/BLMPOP can take multiple keys, it eventually operates on just one key.
And it will propagate as LPOP or RPOP with the COUNT option.
As a new command, it still return NIL if we can't pop any elements.
For the normal response is nested arrays in RESP2 and RESP3, like:
```
LMPOP/BLMPOP
1) keyname
2) 1) element1
2) element2
```
I.e. unlike BLPOP that returns a key name and one element so it uses a flat array,
and LPOP that returns multiple elements with no key name, and again uses a flat array,
this one has to return a nested array, and it does for for both RESP2 and RESP3 (like SCAN does)
Some discuss can see: #766#8824
Add two INFO metrics:
```
total_active_defrag_time:12345
current_active_defrag_time:456
```
`current_active_defrag_time` if greater than 0, means how much time has
passed since active defrag started running. If active defrag stops, this metric is reset to 0.
`total_active_defrag_time` means total time the fragmentation
was over the defrag threshold since the server started.
This is a followup PR for #9031
* Enhance dict to support arbitrary metadata carried in dictEntry
Co-authored-by: Viktor Söderqvist <viktor.soderqvist@est.tech>
* Rewrite slot-to-keys mapping to linked lists using dict entry metadata
This is a memory enhancement for Redis Cluster.
The radix tree slots_to_keys (which duplicates all key names prefixed with their
slot number) is replaced with a linked list for each slot. The dict entries of
the same cluster slot form a linked list and the pointers are stored as metadata
in each dict entry of the main DB dict.
This commit also moves the slot-to-key API from db.c to cluster.c.
Co-authored-by: Jim Brunner <brunnerj@amazon.com>
We implement incremental data sync in rio.c by call fsync, on slow disk, that may cost a lot of time,
sync_file_range could provide async fsync, so we could serialize key/value and sync file data at the same time.
> one tip for sync_file_range usage: http://lkml.iu.edu/hypermail/linux/kernel/1005.2/01845.html
Additionally, this change avoids a single large write to be used, which can result in a mass of dirty
pages in the kernel (increasing the risk of someone else's write to block).
On HDD, current solution could reduce approximate half of dumping RDB time,
this PR costs 50s for dump 7.7G rdb but unstable branch costs 93s.
On NVME SSD, this PR can't reduce much time, this PR costs 40s, unstable branch costs 48s.
Moreover, I find calling data sync every 4MB is better than 32MB.
Part one of implementing #8702 (taking hashes first before other types)
## Description of the feature
1. Change ziplist encoded hash objects to listpack encoding.
2. Convert existing ziplists on RDB loading time. an O(n) operation.
## Rdb format changes
1. Add RDB_TYPE_HASH_LISTPACK rdb type.
2. Bump RDB_VERSION to 10
## Interface changes
1. New `hash-max-listpack-entries` config is an alias for `hash-max-ziplist-entries` (same with `hash-max-listpack-value`)
2. OBJECT ENCODING will return `listpack` instead of `ziplist`
## Listpack improvements:
1. Support direct insert, replace integer element (rather than convert back and forth from string)
3. Add more listpack capabilities to match the ziplist ones (like `lpFind`, `lpRandomPairs` and such)
4. Optimize element length fetching, avoid multiple calculations
5. Use inline to avoid function call overhead.
## Tests
1. Add a new test to the RDB load time conversion
2. Adding the listpack unit tests. (based on the one in ziplist.c)
3. Add a few "corrupt payload: fuzzer findings" tests, and slightly modify existing ones.
Co-authored-by: Oran Agra <oran@redislabs.com>
Recently we found two issues in the fuzzer tester: #9302#9285
After fixing them, more problems surfaced and this PR (as well as #9297) aims to fix them.
Here's a list of the fixes
- Prevent an overflow when allocating a dict hashtable
- Prevent OOM when attempting to allocate a huge string
- Prevent a few invalid accesses in listpack
- Improve sanitization of listpack first entry
- Validate integrity of stream consumer groups PEL
- Validate integrity of stream listpack entry IDs
- Validate ziplist tail followed by extra data which start with 0xff
Co-authored-by: sundb <sundbcn@gmail.com>
Reduce dict struct memory overhead
on 64bit dict size goes down from jemalloc's 96 byte bin to its 56 byte bin.
summary of changes:
- Remove `privdata` from callbacks and dict creation. (this affects many files, see "Interface change" below).
- Meld `dictht` struct into the `dict` struct to eliminate struct padding. (this affects just dict.c and defrag.c)
- Eliminate the `sizemask` field, can be calculated from size when needed.
- Convert the `size` field into `size_exp` (exponent), utilizes one byte instead of 8.
Interface change: pass dict pointer to dict type call back functions.
This is instead of passing the removed privdata field. In the future if
we'd like to have private data in the callbacks we can extract it from
the dict type. We can extend dictType to include a custom dict struct
allocator and use it to allocate more data at the end of the dict
struct. This data can then be used to store private data later acccessed
by the callbacks.
## Backgroud
As we know, after `fork`, one process will copy pages when writing data to these
pages(CoW), and another process still keep old pages, they totally cost more memory.
For redis, we suffered that redis consumed much memory when the fork child is serializing
key/values, even that maybe cause OOM.
But actually we find, in redis fork child process, the child process don't need to keep some
memory and parent process may write or update that, for example, child process will never
access the key-value that is serialized but users may update it in parent process.
So we think it may reduce COW if the child process release memory that it is not needed.
## Implementation
For releasing key value in child process, we may think we call `decrRefCount` to free memory,
but i find the fork child process still use much memory when we don't write any data to redis,
and it costs much more time that slows down bgsave. Maybe because memory allocator doesn't
really release memory to OS, and it may modify some inner data for this free operation, especially
when we free small objects.
Moreover, CoW is based on pages, so it is a easy way that we only free the memory bulk that is
not less than kernel page size. madvise(MADV_DONTNEED) can quickly release specified region
pages to OS bypassing memory allocator, and allocator still consider that this memory still is used
and don't change its inner data.
There are some buffers we can release in the fork child process:
- **Serialized key-values**
the fork child process never access serialized key-values, so we try to free them.
Because we only can release big bulk memory, and it is time consumed to iterate all
items/members/fields/entries of complex data type. So we decide to iterate them and
try to release them only when their average size of item/member/field/entry is more
than page size of OS.
- **Replication backlog**
Because replication backlog is a cycle buffer, it will be changed quickly if redis has heavy
write traffic, but in fork child process, we don't need to access that.
- **Client buffers**
If clients have requests during having the fork child process, clients' buffer also be changed
frequently. The memory includes client query buffer, output buffer, and client struct used memory.
To get child process peak private dirty memory, we need to count peak memory instead
of last used memory, because the child process may continue to release memory (since
COW used to only grow till now, the last was equivalent to the peak).
Also we're adding a new `current_cow_peak` info variable (to complement the existing
`current_cow_size`)
Co-authored-by: Oran Agra <oran@redislabs.com>
## Current state
1. Lua has its own parser that handles parsing `reds.call` replies and translates them
to Lua objects that can be used by the user Lua code. The parser partially handles
resp3 (missing big number, verbatim, attribute, ...)
2. Modules have their own parser that handles parsing `RM_Call` replies and translates
them to RedisModuleCallReply objects. The parser does not support resp3.
In addition, in the future, we want to add Redis Function (#8693) that will probably
support more languages. At some point maintaining so many parsers will stop
scaling (bug fixes and protocol changes will need to be applied on all of them).
We will probably end up with different parsers that support different parts of the
resp protocol (like we already have today with Lua and modules)
## PR Changes
This PR attempt to unified the reply parsing of Lua and modules (and in the future
Redis Function) by introducing a new parser unit (`resp_parser.c`). The new parser
handles parsing the reply and calls different callbacks to allow the users (another
unit that uses the parser, i.e, Lua, modules, or Redis Function) to analyze the reply.
### Lua API Additions
The code that handles reply parsing on `scripting.c` was removed. Instead, it uses
the resp_parser to parse and create a Lua object out of the reply. As mentioned
above the Lua parser did not handle parsing big numbers, verbatim, and attribute.
The new parser can handle those and so Lua also gets it for free.
Those are translated to Lua objects in the following way:
1. Big Number - Lua table `{'big_number':'<str representation for big number>'}`
2. Verbatim - Lua table `{'verbatim_string':{'format':'<verbatim format>', 'string':'<verbatim string value>'}}`
3. Attribute - currently ignored and not expose to the Lua parser, another issue will be open to decide how to expose it.
Tests were added to check resp3 reply parsing on Lua
### Modules API Additions
The reply parsing code on `module.c` was also removed and the new resp_parser is used instead.
In addition, the RedisModuleCallReply was also extracted to a separate unit located on `call_reply.c`
(in the future, this unit will also be used by Redis Function). A nice side effect of unified parsing is
that modules now also support resp3. Resp3 can be enabled by giving `3` as a parameter to the
fmt argument of `RM_Call`. It is also possible to give `0`, which will indicate an auto mode. i.e, Redis
will automatically chose the reply protocol base on the current client set on the RedisModuleCtx
(this mode will mostly be used when the module want to pass the reply to the client as is).
In addition, the following RedisModuleAPI were added to allow analyzing resp3 replies:
* New RedisModuleCallReply types:
* `REDISMODULE_REPLY_MAP`
* `REDISMODULE_REPLY_SET`
* `REDISMODULE_REPLY_BOOL`
* `REDISMODULE_REPLY_DOUBLE`
* `REDISMODULE_REPLY_BIG_NUMBER`
* `REDISMODULE_REPLY_VERBATIM_STRING`
* `REDISMODULE_REPLY_ATTRIBUTE`
* New RedisModuleAPI:
* `RedisModule_CallReplyDouble` - getting double value from resp3 double reply
* `RedisModule_CallReplyBool` - getting boolean value from resp3 boolean reply
* `RedisModule_CallReplyBigNumber` - getting big number value from resp3 big number reply
* `RedisModule_CallReplyVerbatim` - getting format and value from resp3 verbatim reply
* `RedisModule_CallReplySetElement` - getting element from resp3 set reply
* `RedisModule_CallReplyMapElement` - getting key and value from resp3 map reply
* `RedisModule_CallReplyAttribute` - getting a reply attribute
* `RedisModule_CallReplyAttributeElement` - getting key and value from resp3 attribute reply
* New context flags:
* `REDISMODULE_CTX_FLAGS_RESP3` - indicate that the client is using resp3
Tests were added to check the new RedisModuleAPI
### Modules API Changes
* RM_ReplyWithCallReply might return REDISMODULE_ERR if the given CallReply is in resp3
but the client expects resp2. This is not a breaking change because in order to get a resp3
CallReply one needs to specifically specify `3` as a parameter to the fmt argument of
`RM_Call` (as mentioned above).
Tests were added to check this change
### More small Additions
* Added `debug set-disable-deny-scripts` that allows to turn on and off the commands no-script
flag protection. This is used by the Lua resp3 tests so it will be possible to run `debug protocol`
and check the resp3 parsing code.
Co-authored-by: Oran Agra <oran@redislabs.com>
Co-authored-by: Yossi Gottlieb <yossigo@gmail.com>
Add SINTERCARD and ZINTERCARD commands that are similar to
ZINTER and SINTER but only return the cardinality with minimum
processing and memory overheads.
Co-authored-by: Oran Agra <oran@redislabs.com>
Add two INFO metrics:
```
total_eviction_exceeded_time:69734
current_eviction_exceeded_time:10230
```
`current_eviction_exceeded_time` if greater than 0, means how much time current used memory is greater than `maxmemory`. And we are still over the maxmemory. If used memory is below `maxmemory`, this metric is reset to 0.
`total_eviction_exceeded_time` means total time used memory is greater than `maxmemory` since server startup.
The units of these two metrics are ms.
Co-authored-by: Oran Agra <oran@redislabs.com>
GETBIT, SETBIT may access wrong address because of wrap.
BITCOUNT and BITPOS may return wrapped results.
BITFIELD may access the wrong address but also allocate insufficient memory and segfault (see CVE-2021-32761).
This commit uses `uint64_t` or `long long` instead of `size_t`.
related https://github.com/redis/redis/pull/8096
At 32bit platform:
> setbit bit 4294967295 1
(integer) 0
> config set proto-max-bulk-len 536870913
OK
> append bit "\xFF"
(integer) 536870913
> getbit bit 4294967296
(integer) 0
When the bit index is larger than 4294967295, size_t can't hold bit index. In the past, `proto-max-bulk-len` is limit to 536870912, so there is no problem.
After this commit, bit position is stored in `uint64_t` or `long long`. So when `proto-max-bulk-len > 536870912`, 32bit platforms can still be correct.
For 64bit platform, this problem still exists. The major reason is bit pos 8 times of byte pos. When proto-max-bulk-len is very larger, bit pos may overflow.
But at 64bit platform, we don't have so long string. So this bug may never happen.
Additionally this commit add a test cost `512MB` memory which is tag as `large-memory`. Make freebsd ci and valgrind ci ignore this test.
- promote the code in DEBUG PROTOCOL to addReplyBigNum
- DEBUG PROTOCOL ATTRIB skips the attribute when client is RESP2
- networking.c addReply for push and attributes generate assertion when
called on a RESP2 client, anything else would produce a broken
protocol that clients can't handle.
There are two issues fixed in this commit:
1. we want to fail the EXEC command in case there is a watched key that's logically
expired but not yet deleted by active expire or lazy expire.
2. we saw that currently cache time is update in every `call()` (including nested calls),
this time is being also being use for the isKeyExpired comparison, we want to update
the cache time only in the first call (execCommand)
Co-authored-by: Oran Agra <oran@redislabs.com>
