i.e. don't start the search from scratch hitting the used ones again.
this will also reduce the likelihood of collisions (if there are any
left) by increasing the time until we re-use a port we did use in the
past.
apparently when running tests in parallel (the default of --clients 16),
there's a chance for two tests to use the same port.
specifically, one test might shutdown a master and still have the
replica up, and then another test will re-use the port number of master
for another master, and then that replica will connect to the master of
the other test.
this can cause a master to count too many full syncs and fail a test if
we run the tests with --single integration/psync2 --loop --stop
see Probmem 2 in #7314
There's a rare case which leads to stagnation in the defragger, causing
it to keep scanning the keyspace and do nothing (not moving any
allocation), this happens when all the allocator slabs of a certain bin
have the same % utilization, but the slab from which new allocations are
made have a lower utilization.
this commit fixes it by removing the current slab from the overall
average utilization of the bin, and also eliminate any precision loss in
the utilization calculation and move the decision about the defrag to
reside inside jemalloc.
and also add a test that consistently reproduce this issue.
with the original version of 6.0.0, this test detects an excessive full
sync.
with the fix in 1a7cd2c0e, this test detects memory corruption,
especially when using libc allocator with or without valgrind.
this test which has coverage for varoius flows of diskless master was
failing randomly from time to time.
the failure was:
[err]: diskless all replicas drop during rdb pipe in tests/integration/replication.tcl
log message of '*Diskless rdb transfer, last replica dropped, killing fork child*' not found
what seemed to have happened is that the master didn't detect that all
replicas dropped by the time the replication ended, it thought that one
replica is still connected.
now the test takes a few seconds longer but it seems stable.
This bug was introduced by a recent change in which readQueryFromClient
is using freeClientAsync, and despite the fact that now
freeClientsInAsyncFreeQueue is in beforeSleep, that's not enough since
it's not called during loading in processEventsWhileBlocked.
furthermore, afterSleep was called in that case but beforeSleep wasn't.
This bug also caused slowness sine the level-triggered mode of epoll
kept signaling these connections as readable causing us to keep doing
connRead again and again for ll of these, which keep accumulating.
now both before and after sleep are called, but not all of their actions
are performed during loading, some are only reserved for the main loop.
fixes issue #7215
Seems like on some systems choosing specific TLS v1/v1.1 versions no
longer works as expected. Test is reduced for v1.2 now which is still
good enough to test the mechansim, and matters most anyway.
* fix memlry leaks with diskless replica short read.
* fix a few timing issues with valgrind runs
* fix issue with valgrind and watchdog schedule signal
about the valgrind WD issue:
the stack trace test in logging.tcl, has issues with valgrind:
==28808== Can't extend stack to 0x1ffeffdb38 during signal delivery for thread 1:
==28808== too small or bad protection modes
it seems to be some valgrind bug with SA_ONSTACK.
SA_ONSTACK seems unneeded since WD is not recursive (SA_NODEFER was removed),
also, not sure if it's even valid without a call to sigaltstack()
Currently, there are several types of threads/child processes of a
redis server. Sometimes we need deeply optimise the performance of
redis, so we would like to isolate threads/processes.
There were some discussion about cpu affinity cases in the issue:
https://github.com/antirez/redis/issues/2863
So implement cpu affinity setting by redis.conf in this patch, then
we can config server_cpulist/bio_cpulist/aof_rewrite_cpulist/
bgsave_cpulist by cpu list.
Examples of cpulist in redis.conf:
server_cpulist 0-7:2 means cpu affinity 0,2,4,6
bio_cpulist 1,3 means cpu affinity 1,3
aof_rewrite_cpulist 8-11 means cpu affinity 8,9,10,11
bgsave_cpulist 1,10-11 means cpu affinity 1,10,11
Test on linux/freebsd, both work fine.
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Now both master and replicas keep track of the last replication offset
that contains meaningful data (ignoring the tailing pings), and both
trim that tail from the replication backlog, and the offset with which
they try to use for psync.
the implication is that if someone missed some pings, or even have
excessive pings that the promoted replica has, it'll still be able to
psync (avoid full sync).
the downside (which was already committed) is that replicas running old
code may fail to psync, since the promoted replica trims pings form it's
backlog.
This commit adds a test that reproduces several cases of promotions and
demotions with stale and non-stale pings
Background:
The mearningful offset on the master was added recently to solve a problem were
the master is left all alone, injecting PINGs into it's backlog when no one is
listening and then gets demoted and tries to replicate from a replica that didn't
have any of the PINGs (or at least not the last ones).
however, consider this case:
master A has two replicas (B and C) replicating directly from it.
there's no traffic at all, and also no network issues, just many pings in the
tail of the backlog. now B gets promoted, A becomes a replica of B, and C
remains a replica of A. when A gets demoted, it trims the pings from its
backlog, and successfully replicate from B. however, C is still aware of
these PINGs, when it'll disconnect and re-connect to A, it'll ask for something
that's not in the backlog anymore (since A trimmed the tail of it's backlog),
and be forced to do a full sync (something it didn't have to do before the
meaningful offset fix).
Besides that, the psync2 test was always failing randomly here and there, it
turns out the reason were PINGs. Investigating it shows the following scenario:
cycle 1: redis #1 is master, and all the rest are direct replicas of #1
cycle 2: redis #2 is promoted to master, #1 is a replica of #2 and #3 is replica of #1
now we see that when #1 is demoted it prints:
17339:S 21 Apr 2020 11:16:38.523 * Using the meaningful offset 3929963 instead of 3929977 to exclude the final PINGs (14 bytes difference)
17339:S 21 Apr 2020 11:16:39.391 * Trying a partial resynchronization (request e2b3f8817735fdfe5fa4626766daa938b61419e5:3929964).
17339:S 21 Apr 2020 11:16:39.392 * Successful partial resynchronization with master.
and when #3 connects to the demoted #2, #2 says:
17339:S 21 Apr 2020 11:16:40.084 * Partial resynchronization not accepted: Requested offset for secondary ID was 3929978, but I can reply up to 3929964
so the issue here is that the meaningful offset feature saved the day for the
demoted master (since it needs to sync from a replica that didn't get the last
ping), but it didn't help one of the other replicas which did get the last ping.
STRALGO should be a container for mostly read-only string
algorithms in Redis. The algorithms should have two main
characteristics:
1. They should be non trivial to compute, and often not part of
programming language standard libraries.
2. They should be fast enough that it is a good idea to have optimized C
implementations.
Next thing I would love to see? A small strings compression algorithm.
this test is time sensitive and it sometimes fail to pass below the
latency threshold, even on strong machines.
this test was the reson we're running just 2 parallel tests in the
github actions CI, revering this.
There is an inherent race between the deferring client and the
"main" client of the test: While the deferring client issues a blocking
command, we can't know for sure that by the time the "main" client
tries to issue another command (Usually one that unblocks the deferring
client) the deferring client is even blocked...
For lack of a better choice this commit uses TCL's 'after' in order
to give some time for the deferring client to issues its blocking
command before the "main" client does its thing.
This problem probably exists in many other tests but this commit
tries to fix blockonkeys.tcl
By using a "circular BRPOPLPUSH"-like scenario it was
possible the get the same client on db->blocking_keys
twice (See comment in moduleTryServeClientBlockedOnKey)
The fix was actually already implememnted in
moduleTryServeClientBlockedOnKey but it had a bug:
the funxction should return 0 or 1 (not OK or ERR)
Other changes:
1. Added two commands to blockonkeys.c test module (To
reproduce the case described above)
2. Simplify blockonkeys.c in order to make testing easier
3. cast raxSize() to avoid warning with format spec
Makse sure call() doesn't wrap replicated commands with
a redundant MULTI/EXEC
Other, unrelated changes:
1. Formatting compiler warning in INFO CLIENTS
2. Use CLIENT_ID_AOF instead of UINT64_MAX
37a10cef introduced automatic wrapping of MULTI/EXEC for the
alsoPropagate API. However this collides with the built-in mechanism
already present in module.c. To avoid complex changes near Redis 6 GA
this commit introduces the ability to exclude call() MUTLI/EXEC wrapping
for also propagate in order to continue to use the old code paths in
module.c.
First, we must parse the IDs, so that we abort ASAP.
The return value of this command cannot be an error if
the client successfully acknowledged some messages,
so it should be executed in a "all or nothing" fashion.
the AOF will be loaded successfully, but the stream will be missing,
i.e inconsistencies with the original db.
this was because XADD with id of 0-0 would error.
add a test to reproduce.
Redis refusing to run MULTI or EXEC during script timeout may cause partial
transactions to run.
1) if the client sends MULTI+commands+EXEC in pipeline without waiting for
response, but these arrive to the shards partially while there's a busy script,
and partially after it eventually finishes: we'll end up running only part of
the transaction (since multi was ignored, and exec would fail).
2) similar to the above if EXEC arrives during busy script, it'll be ignored and
the client state remains in a transaction.
the 3rd test which i added for a case where MULTI and EXEC are ok, and
only the body arrives during busy script was already handled correctly
since processCommand calls flagTransaction
*** [err]: PSYNC2: total sum of full synchronizations is exactly 4 in tests/integration/psync2.tcl
Expected 5 == 4 (context: type eval line 6 cmd {assert {$sum == 4}} proc ::test)
issue was that sometime the test got an unexpected full sync since it
tried to switch to the replica before it was in sync with it's master.
it seems that running two clients at a time is ok too, resuces action
time from 20 minutes to 10. we'll use this for now, and if one day it
won't be enough we'll have to run just the sensitive tests one by one
separately from the others.
this commit also fixes an issue with the defrag test that appears to be
very rare.
seems that github actions are slow, using just one client to reduce
false positives.
also adding verbose, testing only on latest ubuntu, and building on
older one.
when doing that, i can reduce the test threshold back to something saner
I saw that the new defag test for list was failing in CI recently, so i
reduce it's threshold from 12 to 60.
besides that, i add / improve the latency test for that other two defrag
tests (add a sensitive latency and digest / save checks)
and fix bad usage of debug populate (can't overrides existing keys).
this was the original intention, which creates higher fragmentation.
When active defrag kicks in and finds a big list, it will create a bookmark to
a node so that it is able to resume iteration from that node later.
The quicklist manages that bookmark, and updates it in case that node is deleted.
This will increase memory usage only on lists of over 1000 (see
active-defrag-max-scan-fields) quicklist nodes (1000 ziplists, not 1000 items)
by 16 bytes.
In 32 bit build, this change reduces the maximum effective config of
list-compress-depth and list-max-ziplist-size (from 32767 to 8191)
This bug affected RM_StringToLongDouble and HINCRBYFLOAT.
I added tests for both cases.
Main changes:
1. Fixed string2ld to fail if string contains \0 in the middle
2. Use string2ld in getLongDoubleFromObject - No point of
having duplicated code here
The two changes above broke RM_SaveLongDouble/RM_LoadLongDouble
because the long double string was saved with length+1 (An innocent
mistake, but it's actually a bug - The length passed to
RM_SaveLongDouble should not include the last \0).
If a blocked module client times-out (or disconnects, unblocked
by CLIENT command, etc.) we need to call moduleUnblockClient
in order to free memory allocated by the module sub-system
and blocked-client private data
Other changes:
Made blockedonkeys.tcl tests a bit more aggressive in order
to smoke-out potential memory leaks
This commit solves the following bug:
127.0.0.1:6379> XGROUP CREATE x grp $ MKSTREAM
OK
127.0.0.1:6379> XADD x 666 f v
"666-0"
127.0.0.1:6379> XREADGROUP GROUP grp Alice BLOCK 0 STREAMS x >
1) 1) "x"
2) 1) 1) "666-0"
2) 1) "f"
2) "v"
127.0.0.1:6379> XADD x 667 f v
"667-0"
127.0.0.1:6379> XDEL x 667
(integer) 1
127.0.0.1:6379> XREADGROUP GROUP grp Alice BLOCK 0 STREAMS x >
1) 1) "x"
2) (empty array)
The root cause is that we use s->last_id in streamCompareID
while we should use the last *valid* ID
- make lua-replicate-commands mutable (it never was, but i don't see why)
- make tcp-backlog immutable (fix a recent refactory mistake)
- increase the max limit of a few configs to match what they were before
the recent refactory
This commit solves several edge cases that are related to
exhausting the streamID limits: We should correctly calculate
the succeeding streamID instead of blindly incrementing 'seq'
This affects both XREAD and XADD.
Other (unrelated) changes:
Reply with a better error message when trying to add an entry
to a stream that has exhausted last_id
With the previous API, a NULL return value was ambiguous and could
represent either an old value of NULL or an error condition. The new API
returns a status code and allows the old value to be returned
by-reference.
This commit also includes test coverage based on
tests/modules/datatype.c which did not exist at the time of the original
commit.
it seems that commit b087dd1db6 accidentially changed
gen_write_load to not use deferred client, which causes them to be slower and not
generate high load which they should, making some tests less effecitive
Changes in behavior:
- Change server.stream_node_max_entries from int64_t to long long, so that it can be used by the generic infra
- standard error reply instead of "repl-backlog-size must be 1 or greater" and such
- tls-port and a few TLS booleans were readable (config get) even when USE_OPENSSL was off (now they aren't)
- syslog-enabled, syslog-ident, cluster-enabled, appendfilename, and supervised didn't have a get (now they do)
- pidfile was initialized to NULL in InitServerConfig but had CONFIG_DEFAULT_PID_FILE in rewriteConfig (so the real default was "", but rewrite would cause it to be set), fixed the rewrite.
- TLS config in server.h was uninitialized (if no tls config args were provided)
Adding test for sanity and coverage
there were two lssues, one is taht BGREWRITEAOF failed since the initial one was still in progress
the solution for this one is to enable appendonly from the server startup so there's no initial aofrw.
the other problem was 0 loading progress events, theory is that on some
platforms a sleep of 1 will cause a much greater delay due to the context
switch, but on other platform it doesn't. in theory a sleep of 100 micro
for 1k keys whould take 100ms, and with hz of 500 we should be gettering
50 events (one every 2ms). in practise it doesn't work like that, so trying
to find a sleep that would be long enough but still not cause the test to take
too long.
Calling XADD with 0-0 or 0 would result in creating an
empty key and storing it in the database.
Even worse, because XADD will reply with error the action
will not be replicated, creating a master-replica
inconsistency
- Adding RM_ScanKey
- Adding tests for RM_ScanKey
- Refactoring RM_Scan API
Changes in RM_Scan
- cleanup in docs and coding convention
- Moving out of experimantal Api
- Adding ctx to scan callback
- Dont use cursor of -1 as an indication of done (can be a valid cursor)
- Set errno when returning 0 for various reasons
- Rename Cursor to ScanCursor
- Test filters key that are not strings, and opens a key if NULL
The implementation expose the following new functions:
1. RedisModule_CursorCreate - allow to create a new cursor object for
keys scanning
2. RedisModule_CursorRestart - restart an existing cursor to restart the
scan
3. RedisModule_CursorDestroy - destroy an existing cursor
4. RedisModule_Scan - scan keys
The RedisModule_Scan function gets a cursor object, a callback and void*
(used as user private data).
The callback will be called for each key in the database proving the key
name and the value as RedisModuleKey.
- the API name was odd, separated to two apis one for LRU and one for LFU
- the LRU idle time was in 1 second resolution, which might be ok for RDB
and RESTORE, but i think modules may need higher resolution
- adding tests for LFU and for handling maxmemory policy mismatch
recently added more reads into that function, if a later read fails, i must
either free what's already allocated, or return the pointer so that the free
callback will release it.
Add two new functions that leverage the RedisModuleDataType mechanism
for RDB serialization/deserialization and make it possible to use it
to/from arbitrary strings:
* RM_SaveDataTypeToString()
* RM_LoadDataTypeFromString()
rename RM_ServerInfoGetFieldNumerical RM_ServerInfoGetFieldSigned
move string2ull to util.c
fix leak in RM_GetServerInfo when duplicate info fields exist
looks like each platform implements long double differently (different bit count)
so we can't save them as binary, and we also want to avoid creating a new RDB
format version, so we save these are hex strings using "%La".
This commit includes a change in the arguments of ld2string to support this.
as well as tests for coverage and short reads.
coded by @guybe7
- Add RM_GetServerInfo and friends
- Add auto memory for new opaque struct
- Add tests for new APIs
other minor fixes:
- add const in various char pointers
- requested_section in modulesCollectInfo was actually not sds but char*
- extract new string2d out of getDoubleFromObject for code reuse
Add module API for
* replication hooks: role change, master link status, replica online/offline
* persistence hooks: saving, loading, loading progress
* misc hooks: cron loop, shutdown, module loaded/unloaded
* change the way hooks test work, and add tests for all of the above
startLoading() now gets flag indicating what is loaded.
stopLoading() now gets an indication of success or failure.
adding startSaving() and stopSaving() with similar args and role.
sometimes we have several assertions with the same condition in the same test
at different stages, and when these fail (the ones that print the condition
text) you don't know which one it was. other assertions didn't print the
condition text (variable names), just the expected and unexpected values.
So now, all assertions print context line, and conditin text.
besides, one of the major differences between 'assert' and 'assert_equal',
is that the later is able to print the value that doesn't match the expected.
if there is a rare non-reproducible failure, it is helpful to know what was
the value the test encountered and how far it was from the threshold.
So now, adding assert_lessthan and assert_range that can be used in some places.
were we used just 'assert { a > b }' so far.
Adding a test for coverage for RM_Call in a new "misc" unit
to be used for various short simple tests
also solves compilation warnings in redismodule.h and fork.c
misc:
- handle SSL_has_pending by iterating though these in beforeSleep, and setting timeout of 0 to aeProcessEvents
- fix issue with epoll signaling EPOLLHUP and EPOLLERR only to the write handlers. (needed to detect the rdb pipe was closed)
- add key-load-delay config for testing
- trim connShutdown which is no longer needed
- rioFdsetWrite -> rioFdWrite - simplified since there's no longer need to write to multiple FDs
- don't detect rdb child exited (don't call wait3) until we detect the pipe is closed
- Cleanup bad optimization from rio.c, add another one
* Introduce a connection abstraction layer for all socket operations and
integrate it across the code base.
* Provide an optional TLS connections implementation based on OpenSSL.
* Pull a newer version of hiredis with TLS support.
* Tests, redis-cli updates for TLS support.
now that replica can read rdb directly from the socket, it should avoid exiting
on short read and instead try to re-sync.
this commit tries to have minimal effects on non-diskless rdb reading.
and includes a test that tries to trigger this scenario on various read cases.
* create module API for forking child processes.
* refactor duplicate code around creating and tracking forks by AOF and RDB.
* child processes listen to SIGUSR1 and dies exitFromChild in order to
eliminate a valgrind warning of unhandled signal.
* note that BGSAVE error reply has changed.
valgrind error is:
Process terminating with default action of signal 10 (SIGUSR1)
The implementation of the diskless replication was currently diskless only on the master side.
The slave side was still storing the received rdb file to the disk before loading it back in and parsing it.
This commit adds two modes to load rdb directly from socket:
1) when-empty
2) using "swapdb"
the third mode of using diskless slave by flushdb is risky and currently not included.
other changes:
--------------
distinguish between aof configuration and state so that we can re-enable aof only when sync eventually
succeeds (and not when exiting from readSyncBulkPayload after a failed attempt)
also a CONFIG GET and INFO during rdb loading would have lied
When loading rdb from the network, don't kill the server on short read (that can be a network error)
Fix rdb check when performed on preamble AOF
tests:
run replication tests for diskless slave too
make replication test a bit more aggressive
Add test for diskless load swapdb
Add tests to check basic functionality of this optional keyword, and also tested with
a module (redisgraph). Checked quickly with valgrind, no issues.
Copies name the type name canonicalisation code from `typeCommand`, perhaps this would
be better factored out to prevent the two diverging and both needing to be edited to
add new `OBJ_*` types, but this is a little fiddly with C strings.
The [redis-doc](https://github.com/antirez/redis-doc/blob/master/commands.json) repo
will need to be updated with this new arg if accepted.
A quirk to be aware of here is that the GEO commands are backed by zsets not their own
type, so they're not distinguishable from other zsets.
Additionally, for sparse types this has the same behaviour as `MATCH` in that it may
return many empty results before giving something, even for large `COUNT`s.
In fast systems "SLOWLOG RESET" is fast enough to don't be logged even
when the time limit is "1" sometimes. Leading to false positives such
as:
[err]: SLOWLOG - can be disabled in tests/unit/slowlog.tcl
Expected '1' to be equal to '0'
Now clients that are ready to be terminated asynchronously are processed
more often in beforeSleep() instead of being processed in serverCron().
This means that the test will not be able to catch the moment the client
was terminated, also note that the 'omem' figure now changes in big
steps, because of the new client output buffers layout.
So we have to change the test range in order to accomodate for that.
Yet the test is useful enough to be worth taking, even if its precision
is reduced by this commit. Probably if we get more problems, a thing
that makes sense is just to check that the limit is < 200k. That's more
than enough actually.
solving few replication related tests race conditions which fail on slow machines
bugfix in slave buffers test: since the test is executed twice, each time with
a different commands count, the threshold for the delta can't be a constant.
A filter handle is returned and can be used to unregister a filter. In
the future it can also be used to further configure or manipulate the
filter.
Filters are now automatically unregistered when a module unloads.
The XCLAIM docs state the XCLAIM increments the delivery counter for
messages. This PR makes the code match the documentation - which seems
like the desired behaviour - whilst still allowing RETRYCOUNT to be
specified manually.
My understanding of the way streamPropagateXCLAIM() works is that this
change will safely propagate to replicas since retry count is pulled
directly from the streamNACK struct.
Fixes#5194
This way the behavior is very similar to the past one.
This is useful in order to remember the user she probably failed to
configure a password correctly.
Few tests had borderline thresholds that were adjusted.
The slave buffers test had two issues, preventing the slave buffer from growing:
1) the slave didn't necessarily go to sleep on time, or woke up too early,
now using SIGSTOP to make sure it goes to sleep exactly when we want.
2) the master disconnected the slave on timeout
This should be able to find new bugs and regressions about the new
sorted set update function when ZADD is used to update an element
already existing.
The test is able to find the bug fixed at 2f282aee immediately.
* allowing --single to be repeated
* adding --only so that only a specific test inside a unit can be run
* adding --skiptill useful to resume a test that crashed passed the problematic unit.
useful together with --clients 1
* adding --skipfile to use a file containing list of tests names to skip
* printing the names of the tests that are skiped by skipfile or denytags
* adding --config to add config file options from command line
it looks like on slow machines we're getting:
[err]: slave buffer are counted correctly in tests/unit/maxmemory.tcl
Expected condition '$slave_buf > 2*1024*1024' to be true (16914 > 2*1024*1024)
this is a result of the slave waking up too early and eating the
slave buffer before the traffic and the test ends.
on slower machines, the active defrag test tended to fail.
although the fragmentation ratio was below the treshold, the defragger was
still in the middle of a scan cycle.
this commit changes:
- the defragger uses the current fragmentation state, rather than the cache one
that is updated by server cron every 100ms. this actually fixes a bug of
starting one excess scan cycle
- the test lets the defragger use more CPU cycles, in hope that the defrag
will be faster, but also give it more time before we give up.
A) slave buffers didn't count internal fragmentation and sds unused space,
this caused them to induce eviction although we didn't mean for it.
B) slave buffers were consuming about twice the memory of what they actually needed.
- this was mainly due to sdsMakeRoomFor growing to twice as much as needed each time
but networking.c not storing more than 16k (partially fixed recently in 237a38737).
- besides it wasn't able to store half of the new string into one buffer and the
other half into the next (so the above mentioned fix helped mainly for small items).
- lastly, the sds buffers had up to 30% internal fragmentation that was wasted,
consumed but not used.
C) inefficient performance due to starting from a small string and reallocing many times.
what i changed:
- creating dedicated buffers for reply list, counting their size with zmalloc_size
- when creating a new reply node from, preallocate it to at least 16k.
- when appending a new reply to the buffer, first fill all the unused space of the
previous node before starting a new one.
other changes:
- expose mem_not_counted_for_evict info field for the benefit of the test suite
- add a test to make sure slave buffers are counted correctly and that they don't cause eviction
* fail the test (exit code) in case of timeout.
* add --wait-server to allow attaching a debugger
* add --dont-clean to keep log files when tests are done
RESTORE now supports:
1. Setting LRU/LFU
2. Absolute-time TTL
Other related changes:
1. RDB loading will not override LRU bits when RDB file
does not contain the LRU opcode.
2. RDB loading will not set LRU/LFU bits if the server's
maxmemory-policy does not match.
Removing the fix about 50% of the times the test will not be able to
pass cleanly. It's very hard to write a test that will always fail, or
actually, it is possible but then it's likely that it will consistently
pass if we change some random bit, so better to use randomization here.
problems fixed:
* failing to read fragmentation information from jemalloc
* overflow in jemalloc fragmentation hint to the defragger
* test suite not triggering eviction after population
other fixes / improvements:
- LUA script memory isn't taken from zmalloc (taken from libc malloc)
so it can cause high fragmentation ratio to be displayed (which is false)
- there was a problem with "fragmentation" info being calculated from
RSS and used_memory sampled at different times (now sampling them together)
other details:
- adding a few more allocator info fields to INFO and MEMORY commands
- improve defrag test to measure defrag latency of big keys
- increasing the accuracy of the defrag test (by looking at real grag info)
this way we can use an even lower threshold and still avoid false positives
- keep the old (total) "fragmentation" field unchanged, but add new ones for spcific things
- add these the MEMORY DOCTOR command
- deduct LUA memory from the rss in case of non jemalloc allocator (one for which we don't "allocator active/used")
- reduce sampling rate of the rss and allocator info
After checking with the community via Twitter (here:
https://twitter.com/antirez/status/915130876861788161) the verdict was to
use ":". However I later realized, after users lamented the fact that
it's hard to copy IDs just with double click, that this was the reason
why I moved to "." in the first instance. Fortunately "-", that was the
other option with most votes, also gets selected with double click on
most terminal applications on Linux and MacOS.
So my reasoning was:
1) We can't retain "." because it's actually confusing to newcomers, it
looks like a floating number, people may be tricked into thinking they
can order IDs numerically as floats.
2) Moving to a double-click-to-select format is much better. People will
work with such IDs for long time when coding / debugging. Why making now
a choice that will impact this for the next years?
The only other viable option was "-", and that's what I did. Thanks.
getLongLongFromObject calls string2ll which has this line:
/* Return if not all bytes were used. */
so if you pass an sds with 3 characters "1\01" it will fail.
but getLongDoubleFromObject calls strtold, and considers it ok if eptr[0]==`\0`
i.e. if the end of the string found by strtold ends with null terminator
127.0.0.1:6379> set a 1
OK
127.0.0.1:6379> setrange a 2 2
(integer) 3
127.0.0.1:6379> get a
"1\x002"
127.0.0.1:6379> incrbyfloat a 2
"3"
127.0.0.1:6379> get a
"3"
This commit closes issue #3698, at least for now, since the root cause
was not fixed: the bounding box function, for huge radiuses, does not
return a correct bounding box, there are points still within the radius
that are left outside.
So when using GEORADIUS queries with radiuses in the order of 5000 km or
more, it was possible to see, at the edge of the area, certain points
not correctly reported.
Because the bounding box for now was used just as an optimization, and
such huge radiuses are not common, for now the optimization is just
switched off when the radius is near such magnitude.
Three test cases found by the Continuous Integration test were added, so
that we can easily trigger the bug again, both for regression testing
and in order to properly fix it as some point in the future.
Experimentally verified that it can trigger the issue reverting the fix.
At least on my system... Being the bug time/backlog dependant, it is
very hard to tell if this test will be able to trigger the problem
consistently, however even if it triggers the problem once in a while,
we'll see it in the CI environment at http://ci.redis.io.
Apparently 1.4 is too low compared to what you get in certain setups
(including mine). I raised it to 1.55 that hopefully is still enough to
test that the fragmentation went down from 1.7 but without incurring in
issues, however the test setup may be still fragile so certain times this
may lead to false positives again, it's hard to test for these things
in a determinsitic way.
Related to #3786.
And many other related Github issues... all reporting the same problem.
There was probably just not enough backlog in certain unlucky runs.
I'll ask people that can reporduce if they see now this as fixed as
well.
Testing with Solaris C compiler (SunOS 5.11 11.2 sun4v sparc sun4v)
there were issues compiling due to atomicvar.h and running the
tests also failed because of "tail" usage not conform with Solaris
tail implementation. This commit fixes both the issues.
Slow systems like the original Raspberry PI need more time
than 5 seconds to start the script and detect writes.
After fixing the Raspberry PI can pass the unit without issues.
The test now uses more diverse radius sizes, especially sizes near or
greater the whole earth surface are used, that are known to trigger edge
cases. Moreover the PRNG seeding was probably resulting into the same
sequence tested over and over again, now seeding unsing the current unix
time in milliseconds.
Related to #3631.
This actually includes two changes:
1) No newlines to take the master-slave link up when the upstream master
is down. Doing this is dangerous because the sub-slave often is received
replication protocol for an half-command, so can't receive newlines
without desyncing the replication link, even with the code in order to
cancel out the bytes that PSYNC2 was using. Moreover this is probably
also not needed/sane, because anyway the slave can keep serving
requests, and because if it's configured to don't serve stale data, it's
a good idea, actually, to break the link.
2) When a +CONTINUE with a different ID is received, we now break
connection with the sub-slaves: they need to be notified as well. This
was part of the original specification but for some reason it was not
implemented in the code, and was alter found as a PSYNC2 bug in the
integration testing.
This is the PSYNC2 test that helped find issues in the code, and that
still can show a protocol desync from time to time. Work is in progress
in order to find the issue. For now the test is not enabled in "make
test" and must be run manually.
By grepping the continuous integration errors log a number of GEORADIUS
tests failures were detected.
Fortunately when a GEORADIUS failure happens, the test suite logs enough
information in order to reproduce the problem: the PRNG seed,
coordinates and radius of the query.
By reproducing the issues, three different bugs were discovered and
fixed in this commit. This commit also improves the already good
reporting of the fuzzer and adds the failure vectors as regression
tests.
The issues found:
1. We need larger squares around the poles in order to cover the area
requested by the user. There were already checks in order to use a
smaller step (larger squares) but the limit set (+/- 67 degrees) is not
enough in certain edge cases, so 66 is used now.
2. Even near the equator, when the search area center is very near the
edge of the square, the north, south, west or ovest square may not be
able to fully cover the specified radius. Now a test is performed at the
edge of the initial guessed search area, and larger squares are used in
case the test fails.
3. Because of rounding errors between Redis and Tcl, sometimes the test
signaled false positives. This is now addressed.
Whenever possible the original code was improved a bit in other ways. A
debugging example stanza was added in order to make the next debugging
session simpler when the next bug is found.
During the initial handshake with the master a slave will report to have
a very high disconnection time from its master (since technically it was
disconnected since forever, so the current UNIX time in seconds is
reported).
However when the slave is connected again the Sentinel may re-scan the
INFO output again only after 10 seconds, which is a long time. During
this time Sentinels will consider this instance unable to failover, so
a useless delay is introduced.
Actaully this hardly happened in the practice because when a slave's
master is down, the INFO period for slaves changes to 1 second. However
when a manual failover is attempted immediately after adding slaves
(like in the case of the Sentinel unit test), this problem may happen.
This commit changes the INFO period to 1 second even in the case the
slave's master is not down, but the slave reported to be disconnected
from the master (by publishing, last time we checked, a master
disconnection time field in INFO).
This change is required as a result of an unrelated change in the
replication code that adds a small delay in the master-slave first
synchronization.
The test works but is very slow so far, since it involves resharding
1/5 of all the cluster slots from master 0 to the other 4 masters and
back into the original master.
Now elements added to lists are incremental numbers in order to
understand, when inconsistencies are found, what is the order in which
the elements were added. Also the error now provides both the expected
and found value.
This fix, provided by Paul Kulchenko (@pkulchenko), allows the Lua
scripting engine to evaluate statements with a trailing comment like the
following one:
EVAL "print() --comment" 0
Lua can't parse the above if the string does not end with a newline, so
now a final newline is always added automatically. This does not change
the SHA1 of scripts since the SHA1 is computed on the body we pass to
EVAL, without the other code we add to register the function.
Close#2951.
It's a key invariant that when AOF is enabled, after the cluster
reshards, a crash-recovery event causes all the keys to be still fine
with the expected logical content. Now this is part of unit 04.
The old version was modeled with two failovers, however after the first
it is possible that another slave will migrate to the new master, since
for some time the new master is not backed by any slave. Probably there
should be some pause after a failover, before the migration. Anyway the
test is simpler in this way, and depends less on timing.
64 bit double math is not enough to make the test passing, and rounding
to 1.2999999 instead of 1.23 is not an error in the implementation.
Valgrind and sometimes other archs are not able to work with 80 bit
doubles.
An user raised a question about a given behavior of PFCOUNT. Added a
test to show the behavior (union) is correct when most of the items are
in common.
HINCRBY* tests later used the value "tmp" that was sometimes generated
by the random key generation function. The result was ovewriting what
Tcl expected to be inside Redis with another value, causing the next
HSTRLEN test to fail.
Georadius works by computing the center + neighbors squares covering all
the area of the specified position and radius. Then a distance filter is
used to remove elements which are actually outside the range.
When a huge radius is used, like 5000 km or more, adjacent neighbors may
collide and be the same, leading to the reporting of the same element
multiple times. This only happens in the edge case of huge radius but is
not ideal.
A robust but slow solution would involve qsorting the range to remove
all the duplicates. However since the collisions are only in adjacent
boxes, for the way they are ordered in the code, it is much faster to
just check if the current box is the same as the previous one processed.
This commit adds a regression test for the bug.
Fixes#2767.
MOVE was not able to move the TTL: when a key was moved into a different
database number, it became persistent like if PERSIST was used.
In some incredible way (I guess almost nobody uses Redis MOVE) this bug
remained unnoticed inside Redis internals for many years.
Finally Andy Grunwald discovered it and opened an issue.
This commit fixes the bug and adds a regression test.
Close#2765.
This additional info may provide more clues about the test randomly
failing from time to time. Probably the failure is due to some previous
test that overwrites the logical content in the Tcl variable, but this
will make the problem more obvious.
Rationale:
1. The commands look like internals exposed without a real strong use
case.
2. Whatever there is an use case, the client would implement the
commands client side instead of paying RTT just to use a simple to
reimplement library.
3. They add complexity to an otherwise quite straightforward API.
So for now KILLED ;-)
The GIS standard and all the major DBs implementing GIS related
functions take coordinates as x,y that is longitude,latitude.
It was a bad start for Redis to do things differently, so even if this
means that existing users of the Geo module will be required to change
their code, Redis now conforms to the standard.
Usually Redis is very backward compatible, but this is not an exception
to this rule, since this is the first Geo implementation entering the
official Redis source code. It is not wise to try to be backward
compatible with code forks... :-)
Close#2637.
We set random points in the world, pick a random position, and check if
the returned points by Redis match the ones computed by Tcl by brute
forcing all the points using the distance between two points formula.
This approach is sounding since immediately resulted in finding a bug in
the original implementation.
Current todo:
- replace functions in zset.{c,h} with a new unified Redis
zset access API.
Once we get the zset interface fixed, we can squash
relevant commits in this branch and have one nice commit
to merge into unstable.
This commit adds:
- Geo commands
- Tests; runnable with: ./runtest --single unit/geo
- Geo helpers in deps/geohash-int/
- src/geo.{c,h} and src/geojson.{c,h} implementing geo commands
- Updated build configurations to get everything working
- TEMPORARY: src/zset.{c,h} implementing zset score and zset
range reading without writing to client output buffers.
- Modified linkage of one t_zset.c function for use in zset.c
Conflicts:
src/Makefile
src/redis.c
1. HVSTRLEN -> HSTRLEN. It's unlikely one needs the length of the key,
not clear how the API would work (by value does not make sense) and
there will be better names anyway.
2. Default is to return 0 when field is missing.
3. Default is to return 0 when key is missing.
4. The implementation was slower than needed, and produced unnecessary COW.
Related issue #2415.
This test on Linux was extremely slow, since in Tcl we can't enable
easily tcp-nodelay, so the busy loop used to take *a lot* with bigger
writes. Fixed using pipelining.
This removes:
- list-max-ziplist-entries
- list-max-ziplist-value
This adds:
- list-max-ziplist-size
- list-compress-depth
Also updates config file with new sections and updates
tests to use quicklist settings instead of old list settings.
Previously, the old test ran 5,000 loops and used about 500k.
With quicklist, storing those same 5,000 loops takes up 24k, so the
"large value check" failed!
This increases the test to 20,000 loops which makes the object dump 96k.
This replaces individual ziplist vs. linkedlist representations
for Redis list operations.
Big thanks for all the reviews and feedback from everybody in
https://github.com/antirez/redis/pull/2143
spopCommand() now runs spopWithCountCommand() in case the <count> param is found.
Added intsetRandomMembers() to Intset: Copies N random members from the set into inputted 'values' array. Uses either the Knuth or Floyd sample algos depending on ratio count/size.
Added setTypeRandomElements() to SET type: Returns a number of random elements from a non empty set. This is a version of setTypeRandomElement() that is modified in order to return multiple entries, using dictGetRandomKeys() and intsetRandomMembers().
Added tests for SPOP with <count>: unit/type/set, unit/scripting, integration/aof
--
Cleaned up code a bit to match with required Redis coding style
start_server now uses return value from Tcl exec to get the server pid,
however this introduces errors that depend from timing: a lot of the
testing code base assumed the server to be actually up and running when
server_start returns.
So the old code that waits to see the pid in the log file was restored.
Basically: test to make sure we can load cmsgpack
and do some sanity checks to make sure pack/unpack works
properly. We also have a bonus test for circular encoding
and decoding because I was curious how it worked.
People mostly use SORT against lists, but our prior
behavior was pretending lists were an unordered bag
requiring a forced-sort when no sort was requested.
We can just use the native list ordering to ensure
consistency across replicaion and scripting calls.
Closes#2079Closes#545 (again)
A few people have written custom C commands because bit
manipulation isn't exposed through Lua. Let's give
them Mike Pall's bitop.
This adds bitop 1.0.2 (2012-05-08) from http://bitop.luajit.org/
bitop is imported as "bit" into the global namespace.
New Lua commands: bit.tobit, bit.tohex, bit.bnot, bit.band, bit.bor, bit.bxor,
bit.lshift, bit.rshift, bit.arshift, bit.rol, bit.ror, bit.bswap
Verification of working (the asserts would abort on error, so (nil) is correct):
127.0.0.1:6379> eval "assert(bit.tobit(1) == 1); assert(bit.band(1) == 1); assert(bit.bxor(1,2) == 3); assert(bit.bor(1,2,4,8,16,32,64,128) == 255)" 0
(nil)
127.0.0.1:6379> eval 'assert(0x7fffffff == 2147483647, "broken hex literals"); assert(0xffffffff == -1 or 0xffffffff == 2^32-1, "broken hex literals"); assert(tostring(-1) == "-1", "broken tostring()"); assert(tostring(0xffffffff) == "-1" or tostring(0xffffffff) == "4294967295", "broken tostring()")' 0
(nil)
Tests also integrated into the scripting tests and can be run with:
./runtest --single unit/scripting
Tests are excerpted from `bittest.lua` included in the bitop distribution.
When aof-load-truncated option was introduced, with a default of "yes",
the past behavior of the server to abort with trunncated AOF changed, so
we need to explicitly configure the tests to abort with truncated AOF
by setting the option to no.
Previously, "MOVE key somestring" would move the key to
DB 0 which is just unexpected and wrong.
String as DB == error.
Test added too.
Modified by @antirez in order to use the getLongLongFromObject() API
instead of strtol().
Fixes#1428
Also adds test for numsub — due to tcl being tcl,
it doesn't capture the "numberness" of the fix,
but now we at least have one test case for numsub.
Closes#1561
We only want to use the last STORE key, but we have to record
we actually found a STORE key so we can increment the final return
key count.
Test added to prevent further regression.
Closes#1883, #1645, #1647
Previously the end was casted to a smaller type
which resulted in a wrong check and failed
with values larger than handled by unsigned.
Closes#1847, #1844
In the test we use WAIT when the master and slave are up, and only later the
partition is created killing the master, so we are sure we don't incur
in failure modes that may lose writes in this test: the goal here is to
make sure that the elected slave was replicating correctly with the
master.
In the initialization test for each instance we used to unregister the
old master and register it again to clear the config.
However there is a race condition doing this: as soon as we unregister
and re-register "mymaster", another Sentinel can update the new
configuration with the old state because of gossip "hello" messages.
So the correct procedure is instead, unregister "mymaster" from all the
sentinel instances, and re-register it everywhere again.
Lua scripts are executed in the context of the currently selected
database (as selected by the caller of the script).
However Lua scripts are also free to use the SELECT command in order to
affect other DBs. When SELECT is called frm Lua, the old behavior, before
this commit, was to automatically set the Lua caller selected DB to the
last DB selected by Lua. See for example the following sequence of
commands:
SELECT 0
SET x 10
EVAL "redis.call('select','1')" 0
SET x 20
Before this commit after the execution of this sequence of commands,
we'll have x=10 in DB 0, and x=20 in DB 1.
Because of the problem above, there was a bug affecting replication of
Lua scripts, because of the actual implementation of replication. It was
possible to fix the implementation of Lua scripts in order to fix the
issue, but looking closely, the bug is the consequence of the behavior
of Lua ability to set the caller's DB.
Under the old semantics, a script selecting a different DB, has no simple
ways to restore the state and select back the previously selected DB.
Moreover the script auhtor must remember that the restore is needed,
otherwise the new commands executed by the caller, will be executed in
the context of a different DB.
So this commit fixes both the replication issue, and this hard-to-use
semantics, by removing the ability of Lua, after the script execution,
to force the caller to switch to the DB selected by the Lua script.
The new behavior of the previous sequence of commadns is to just set
X=20 in DB 0. However Lua scripts are still capable of writing / reading
from different DBs if needed.
WARNING: This is a semantical change that will break programs that are
conceived to select the client selected DB via Lua scripts.
This fixes issue #1811.
The new check-for-number behavior of Lua arguments broke
users who use large strings of just integers.
The Lua number check would convert the string to a number, but
that breaks user data because
Lua numbers have limited precision compared to an arbitrarily
precise number wrapped in a string.
Regression fixed and new test added.
Fixes#1118 again.
FLUSHALL will fail on read-only slaves, but there the command is not
needed in order to reset the instance with CLUSTER RESET so errors can
be ignored.
Previously the PID format was:
[PID] Timestamp
But it recently changed to:
PID:X Timestamp
The tcl testing framework was grabbing the PID from \[\d+\], but
that's not valid anymore.
Now we grab the pid from "PID: <PID>" in the part of Redis startup
output to the right of the ASCII logo.
The bug was triggered by running the test with Valgrind (which is a lot
slower and more sensible to timing issues) after the recent changes
that made Redis more promptly able to reply with the -LOADING error.
Behrad Zari discovered [1] and Josiah reported [2]: if you block
and wait for a list to exist, but the list creates from
a non-push command, the blocked client never gets notified.
This commit adds notification of blocked clients into
the DB layer and away from individual commands.
Lists can be created by [LR]PUSH, SORT..STORE, RENAME, MOVE,
and RESTORE. Previously, blocked client notifications were
only triggered by [LR]PUSH. Your client would never get
notified if a list were created by SORT..STORE or RENAME or
a RESTORE, etc.
Blocked client notification now happens in one unified place:
- dbAdd() triggers notification when adding a list to the DB
Two new tests are added that fail prior to this commit.
All test pass.
Fixes#1668
[1]: https://groups.google.com/forum/#!topic/redis-db/k4oWfMkN1NU
[2]: #1668
Better handling of connection errors in order to update the table and
recovery, populate the startup nodes table after fetching the list of
nodes.
More work to do about it, it is still not as reliable as
redis-rb-cluster implementation which is the minimal reference
implementation for Redis Cluster clients.
SPOP, tested in the new test, is among the commands rewritng the
client->argv argument vector (it gets rewritten as SREM) for command
replication purposes.
Because of recent optimizations to client->argv caching in the context
of the Lua internal Redis client, it is important to test for SPOP to be
callable from Lua without bad effects to the other commands.
Sometimes the process is still there but no longer in a state that can
be checked (after being killed). This used to happen after a call to
SHUTDOWN NOSAVE in the scripting unit, causing a false positive.
This makes tests a bit slower, but it is better to test things at a
decent scale instead of using just a few nodes, and for a few tests we
actually need so many nodes.
The test now runs in a self-contained directory.
The general abstractions to run the tests in an environment where
mutliple instances are executed at the same time was extrapolated into
instances.tcl, that will be reused to test Redis Cluster.
This commit sets the failover timeout to 30 seconds instead of the 180
seconds default, and allows to reconfigure multiple slaves at the same
time.
This makes tests less sensible to timing, with the result that there are
less false positives due to normal behaviors that require time to
succeed or to be retried.
However the long term solution is probably some way in order to detect
when a test failed because of timing issues (for example split brain
during leader election) and retry it.
It was verified in practice that this test is able to stress much more
the implementation by introducing errors that were only trivially to
detect with different offsets but impossible to detect starting always
at zero and counting bits the full length of the string.
An unit can abort in the middle for an error. The next unit should not
assume that the instances are in a clean state, and must restart what
was left killed.
Sentinel tests are designed to be dependent on the previous tests in the
same unit, so usually we can't continue with the next test in the same
unit if a previous test failed.
The test was previously performed by removing the master from the
Sentinel monitored masters. The test with the Sentinels crashed is
more similar to real-world partitions / failures.
The area a number of mandatory tests to craete a stable setup for
testing that is not too sensitive to timing issues. All those tests
moved to includes/init-tests, and marked as (init).
It is now possible to kill and restart sentinel or redis instances for
more real-world testing.
The 01 unit tests the capability of Sentinel to update the configuration
of Sentinels rejoining the cluster, however the test is pretty trivial
and more tests should be added.
Some inline test moved into server_is_up procedure.
Also find_available_port was moved into util since it is going
to be used for the Sentinel test as well.
The Redis test uses a server-clients model in order to parallelize the
execution of different tests. However in recent versions of osx not
setting the channel to a binary encoding caused issues even if AFAIK no
binary data is really sent via this channel. However now the channels
are deliberately set to a binary encoding and this solves the issue.
The exact issue was the test not terminating and giving the impression
of running forever, since test clients or servers were unable to
exchange the messages to continue.
When the test is executed using the root account, setting the permission
to 222 does not work as expected, as root can read files with 222
permission.
Now we skip the test if root is detected.
This fixes issue #1034 and the duplicated #1040 issue.
Thanks to Jan-Erik Rediger (@badboy on Github) for finding a way to reproduce the issue.
When keyspace events are enabled, the overhead is not sever but
noticeable, so this commit introduces the ability to select subclasses
of events in order to avoid to generate events the user is not
interested in.
The events can be selected using redis.conf or CONFIG SET / GET.
UNSUBSCRIBE and PUNSUBSCRIBE commands are designed to mass-unsubscribe
the client respectively all the channels and patters if called without
arguments.
However when these functions are called without arguments, but there are
no channels or patters we are subscribed to, the old behavior was to
don't reply at all.
This behavior is broken, as every command should always reply.
Also it is possible that we are no longer subscribed to a channels but we
are subscribed to patters or the other way around, and the client should
be notified with the correct number of subscriptions.
Also it is not pretty that sometimes we did not receive a reply at all
in a redis-cli session from these commands, blocking redis-cli trying
to read the reply.
This fixes issue #714.
The Redis Slow Log always used to log the slow commands executed inside
a MULTI/EXEC block. However also EXEC was logged at the end, which is
perfectly useless.
Now EXEC is no longer logged and a test was added to test this behavior.
This fixes issue #759.
Due to changes in recent releases of osx leaks utility, the osx leak
detection no longer worked. Now it is fixed in a way that should be
backward compatible.
SDIFF used an algorithm that was O(N) where N is the total number
of elements of all the sets involved in the operation.
The algorithm worked like that:
ALGORITHM 1:
1) For the first set, add all the members to an auxiliary set.
2) For all the other sets, remove all the members of the set from the
auxiliary set.
So it is an O(N) algorithm where N is the total number of elements in
all the sets involved in the diff operation.
Cristobal Viedma suggested to modify the algorithm to the following:
ALGORITHM 2:
1) Iterate all the elements of the first set.
2) For every element, check if the element also exists in all the other
remaining sets.
3) Add the element to the auxiliary set only if it does not exist in any
of the other sets.
The complexity of this algorithm on the worst case is O(N*M) where N is
the size of the first set and M the total number of sets involved in the
operation.
However when there are elements in common, with this algorithm we stop
the computation for a given element as long as we find a duplicated
element into another set.
I (antirez) added an additional step to algorithm 2 to make it faster,
that is to sort the set to subtract from the biggest to the
smallest, so that it is more likely to find a duplicate in a larger sets
that are checked before the smaller ones.
WHAT IS BETTER?
None of course, for instance if the first set is much larger than the
other sets the second algorithm does a lot more work compared to the
first algorithm.
Similarly if the first set is much smaller than the other sets, the
original algorithm will less work.
So this commit makes Redis able to guess the number of operations
required by each algorithm, and select the best at runtime according
to the input received.
However, since the second algorithm has better constant times and can do
less work if there are duplicated elements, an advantage is given to the
second algorithm.
EVALSHA used to crash if the SHA1 was not lowercase (Issue #783).
Fixed using a case insensitive dictionary type for the sha -> script
map used for replication of scripts.
The previous behavior was to return -1 if:
1) Existing key but without an expire set.
2) Non existing key.
Now the second case is handled in a different, and TTL will return -2
if the key does not exist at all.
PTTL follows the same behavior as well.
With COPY now MIGRATE does not remove the key from the source instance.
With REPLACE it uses RESTORE REPLACE on the target host so that even if
the key already eixsts in the target instance it will be overwritten.
The options can be used together.
The REPLACE option deletes an existing key with the same name (if any)
and materializes the new one. The default behavior without RESTORE is to
return an error if a key already exists.
So instead to reply with a generic error like:
-ERR ... wrong kind of value ...
now it replies with:
-WRONGTYPE ... wrong kind of value ...
This makes this particular error easy to check without resorting to
(fragile) pattern matching of the error string (however the error string
used to be consistent already).
Client libraries should return a specific exeption type for this error.
Most of the commit is about fixing unit tests.
When calling SCRIPT KILL currently you can get two errors:
* No script in timeout (busy) state.
* The script already performed a write.
It is useful to be able to distinguish the two errors, but right now both
start with "ERR" prefix, so string matching (that is fragile) must be used.
This commit introduces two different prefixes.
-NOTBUSY and -UNKILLABLE respectively to reply with an error when no
script is busy at the moment, and when the script already executed a
write operation and can not be killed.
When SORT is called with the option BY set to a string constant not
inclduing the wildcard character "*", there is no way to sort the output
so any ordering is valid. This allows the SORT internals to optimize its
work and don't really sort the output at all.
However it was odd that this option was not able to retain the natural
order of a sorted set. This feature was requested by users multiple
times as sometimes to call SORT with GET against sorted sets as a way to
mass-fetch objects can be handy.
This commit introduces two things:
1) The ability of SORT to return sorted sets elements in their natural
ordering when `BY nosort` is specified, accordingly to `DESC / ASC` options.
2) The ability of SORT to optimize this case further if LIMIT is passed
as well, avoiding to really fetch the whole sorted set, but directly
obtaining the specified range.
Because in this case the sorting is always deterministic, no
post-sorting activity is performed when SORT is called from a Lua
script.
This commit fixes issue #98.
Lua arrays can't contain nil elements (see
http://www.lua.org/pil/19.1.html for more information), so Lua scripts
were not able to return a multi-bulk reply containing nil bulk
elements inside.
This commit introduces a special conversion: a table with just
a "nilbulk" field set to a boolean value is converted by Redis as a nil
bulk reply, but at the same time for Lua this type is not a "nil" so can
be used inside Lua arrays.
This type is also assigned to redis.NIL, so the following two forms
are equivalent and will be able to return a nil bulk reply as second
element of a three elements array:
EVAL "return {1,redis.NIL,3}" 0
EVAL "return {1,{nilbulk=true},3}" 0
The result in redis-cli will be:
1) (integer) 1
2) (nil)
3) (integer) 3
Redis provides support for blocking operations such as BLPOP or BRPOP.
This operations are identical to normal LPOP and RPOP operations as long
as there are elements in the target list, but if the list is empty they
block waiting for new data to arrive to the list.
All the clients blocked waiting for th same list are served in a FIFO
way, so the first that blocked is the first to be served when there is
more data pushed by another client into the list.
The previous implementation of blocking operations was conceived to
serve clients in the context of push operations. For for instance:
1) There is a client "A" blocked on list "foo".
2) The client "B" performs `LPUSH foo somevalue`.
3) The client "A" is served in the context of the "B" LPUSH,
synchronously.
Processing things in a synchronous way was useful as if "A" pushes a
value that is served by "B", from the point of view of the database is a
NOP (no operation) thing, that is, nothing is replicated, nothing is
written in the AOF file, and so forth.
However later we implemented two things:
1) Variadic LPUSH that could add multiple values to a list in the
context of a single call.
2) BRPOPLPUSH that was a version of BRPOP that also provided a "PUSH"
side effect when receiving data.
This forced us to make the synchronous implementation more complex. If
client "B" is waiting for data, and "A" pushes three elemnents in a
single call, we needed to propagate an LPUSH with a missing argument
in the AOF and replication link. We also needed to make sure to
replicate the LPUSH side of BRPOPLPUSH, but only if in turn did not
happened to serve another blocking client into another list ;)
This were complex but with a few of mutually recursive functions
everything worked as expected... until one day we introduced scripting
in Redis.
Scripting + synchronous blocking operations = Issue #614.
Basically you can't "rewrite" a script to have just a partial effect on
the replicas and AOF file if the script happened to serve a few blocked
clients.
The solution to all this problems, implemented by this commit, is to
change the way we serve blocked clients. Instead of serving the blocked
clients synchronously, in the context of the command performing the PUSH
operation, it is now an asynchronous and iterative process:
1) If a key that has clients blocked waiting for data is the subject of
a list push operation, We simply mark keys as "ready" and put it into a
queue.
2) Every command pushing stuff on lists, as a variadic LPUSH, a script,
or whatever it is, is replicated verbatim without any rewriting.
3) Every time a Redis command, a MULTI/EXEC block, or a script,
completed its execution, we run the list of keys ready to serve blocked
clients (as more data arrived), and process this list serving the
blocked clients.
4) As a result of "3" maybe more keys are ready again for other clients
(as a result of BRPOPLPUSH we may have push operations), so we iterate
back to step "3" if it's needed.
The new code has a much simpler semantics, and a simpler to understand
implementation, with the disadvantage of not being able to "optmize out"
a PUSH+BPOP as a No OP.
This commit will be tested with care before the final merge, more tests
will be added likely.
Bug #582 was not present in 32 bit builds of Redis as
getObjectFromLong() will return an error for overflow.
This commit makes sure that the test does not fail because of the error
returned when running against 32 bit builds.
remove unsafe and unnecessary cast.
until now, this cast may lead segmentation fault when end > UINT_MAX
setbit foo 0 1
bitcount 0 4294967295
=> ok
bitcount 0 4294967296
=> cause segmentation fault.
Note by @antirez: the commit was modified a bit to also change the
string length type to long, since it's guaranteed to be at max 512 MB in
size, so we can work with the same type across all the code path.
A regression test was also added.
SORT is able to return (faster than when ordering) unordered output if
the "BY" clause is used with a constant value. However we try to play
well with scripting requirements of determinism providing always sorted
outputs when SORT (and other similar commands) are called by Lua
scripts.
However we used the general mechanism in place in scripting in order to
reorder SORT output, that is, if the command has the "S" flag set, the
Lua scripting engine will take an additional step when converting a
multi bulk reply to Lua value, calling a Lua sorting function.
This is suboptimal as we can do it faster inside SORT itself.
This is also broken as issue #545 shows us: basically when SORT is used
with a constant BY, and additionally also GET is used, the Lua scripting
engine was trying to order the output as a flat array, while it was
actually a list of key-value pairs.
What we do know is to recognized if the caller of SORT is the Lua client
(since we can check this using the REDIS_LUA_CLIENT flag). If so, and if
a "don't sort" condition is triggered by the BY option with a constant
string, we force the lexicographical sorting.
This commit fixes this bug and improves the performance, and at the same
time simplifies the implementation. This does not mean I'm smart today,
it means I was stupid when I committed the original implementation ;)
Redis used to crash with a call like the following:
EVAL "redis.call()" 0
Now the explicit check for at least one argument prevents the problem.
This commit fixes issue #655.
The new fuzzy tester also removes elements from the hash instead of just
adding random fields. This should increase the probability to find bugs
in the implementations of the hash type internal representations.
A new stress test was added to stress test the code converting a ziplist
into an hash table.
In this commit also randomValue helper function was modified to also
return negative values.
wait_for_condition is now used instead of the usual "after 1000" (that
is the way to sleep in Tcl). This should avoid to find the replica in
a state where it is loading the RDB in memory, returning -LOADING error.
This test used to fail when running the test over valgrind, due to the
added latencies.
(additional commit notes by antirez@gmail.com):
The rdbIsObjectType() macro was not updated when the new RDB object type
of ziplist encoded hashes was added.
As a result RESTORE, that uses rdbLoadObjectType(), failed when a
ziplist encoded hash was loaded.
This does not affected normal RDB loading because in that case we use
the lower-level function rdbLoadType().
The commit also adds a regression test.
In the issue #529 an user reported a bug that can be triggered with the
following code:
flushdb
set a
"\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
bitop or x a b
The bug was introduced with the speed optimization in commit 8bbc076
that specializes every BITOP operation loop up to the minimum length of
the input strings.
However the computation of the minimum length contained an error when a
non existing key was present in the input, after a key that was non zero
length.
This commit fixes the bug and adds a regression test for it.
This commit adds a fast-path to the BITOP that can be used for all the
bytes from 0 to the minimal length of the string, and if there are
at max 16 input keys.
Often the intersected bitmaps are roughly the same size, so this
optimization can provide a 10x speed boost to most real world usages
of the command.
Bytes are processed four full words at a time, in loops specialized
for the specific BITOP sub-command, without the need to check for
length issues with the inputs (since we run this algorithm only as far
as there is data from all the keys at the same time).
The remaining part of the string is intersected in the usual way using
the slow but generic algorith.
It is possible to do better than this with inputs that are not roughly
the same size, sorting the input keys by length, by initializing the
result string in a smarter way, and noticing that the final part of the
output string composed of only data from the longest string does not
need any proecessing since AND, OR and XOR against an empty string does
not alter the output (zero in the first case, and the original string in
the other two cases).
More implementations will be implemented later likely, but this should
be enough to release Redis 2.6-RC4 with bitops merged in.
Note: this commit also adds better testing for BITOP NOT command, that
is currently the faster and hard to optimize further since it just
flips the bits of a single input string.
A bug in the implementation caused BITOP to crash the server if at least
one one of the source objects was integer encoded.
The new implementation takes an additional array of Redis objects
pointers and calls getDecodedObject() to get a reference to a string
encoded object, and then uses decrRefCount() to release the object.
Tests modified to cover the regression and improve coverage.
Fuzzing tests of BITCOUNT / BITOP are iterated multiple times.
The new BITCOUNT fuzzing test uses random strings in a wider interval of
lengths including zero-len strings.
The Redis implementation is tested against Tcl implementations of the
same operation. Both fuzzing and testing of specific aspects of the
commands behavior are performed.
Weeks ago trying to fix an harmless GCC warning I introduced a bug in
the ziplist-encoded implementations of sorted sets.
The bug completely broke zuiNext() iterator, that is used in the
ZINTERSTORE and ZUNIONSTORE implementation, so those two commands are no
longer reliable starting from Redis version 2.4.12 and latest 2.6.0-RC
releases.
This commit fixes the problem and adds a regression test.
Due to a change in the format of the bug report in case of crash of
failed assertion the test suite was no longer able to properly log it.
Instead just a protocol error was logged by the Redis TCL client that
provided no clue about the actual problem.
This commit resolves the issue by logging everything from the first line
of the log including the string REDIS BUG REPORT, till the end of the
file.
