13087 Commits

Author SHA1 Message Date
Yossi Gottlieb
74d9d95449 Add language servers stuff, test/tls to gitignore. (#7698) 2020-08-24 12:54:56 +03:00
John Sully
ff9df842d8 Implement use-fork config (fails with diskless repl)
Former-commit-id: f2d5c2bca22e9fd506db123c47b7f60cdded7e2c
2020-08-24 03:17:59 +00:00
John Sully
91bd125fb4 Implement use-fork config (fails with diskless repl)
Former-commit-id: 4e941b8d30612c1051a9dde67effaf8c5a86456f
2020-08-24 03:17:31 +00:00
Valentino Geron
7e6c9ef881 Assert that setDeferredAggregateLen isn't called with negative value
In case the redis is about to return broken reply we want to crash
with assert so that we are notified about the bug. see #7687.
2020-08-23 16:03:30 +03:00
Valentino Geron
7a555da64f Fix LPOS command when RANK is greater than matches
When calling to LPOS command when RANK is higher than matches,
the return value is non valid response. For example:
```
LPUSH l a
:1
LPOS l b RANK 5 COUNT 10
*-4
```
It may break client-side parser.

Now, we count how many replies were replied in the array.
```
LPUSH l a
:1
LPOS l b RANK 5 COUNT 10
*0
```
2020-08-23 16:03:30 +03:00
Yossi Gottlieb
257f9f462f Tests: fix redis-cli with remote hosts. (#7693) 2020-08-23 10:17:43 +03:00
Wen Hui
7386b998e8 fix make warnings (#7692) 2020-08-21 23:37:49 +03:00
Wen Hui
86cd4629ae use dictSlots for getting total slots number in dict (#7691) 2020-08-21 00:14:09 +03:00
huangzhw
b980e99929 RedisModuleEvent_LoadingProgress always at 100% progress (#7685)
It was also using the wrong struct, but luckily RedisModuleFlushInfo and RedisModuleLoadingProgress
are identical.
2020-08-20 23:31:06 +03:00
guybe7
4de17eb032 Modules: Invalidate saved_oparray after use (#7688)
We wanna avoid a chance of someone using the pointer in it after it'll be freed / realloced.
2020-08-20 19:55:14 +03:00
杨博东
5e6212e087 Fix flock cluster config may cause failure to restart after kill -9 (#7674)
After fork, the child process(redis-aof-rewrite) will get the fd opened
by the parent process(redis), when redis killed by kill -9, it will not
graceful exit(call prepareForShutdown()), so redis-aof-rewrite thread may still
alive, the fd(lock) will still be held by redis-aof-rewrite thread, and
redis restart will fail to get lock, means fail to start.

This issue was causing failures in the cluster tests in github actions.

Co-authored-by: Oran Agra <oran@redislabs.com>
2020-08-20 08:59:02 +03:00
Raghav Muddur
f7cf24b5da Update clusterMsgDataPublish to clusterMsgModule (#7682)
Correcting the variable to clusterMsgModule.
2020-08-19 19:13:32 -07:00
Madelyn Olson
17c6b3c7ee Fixed hset error since it's shared with hmset (#7678) 2020-08-19 19:07:43 -07:00
Wang Yuan
77155c4744 Add comments on 'slave.repldboff' when use diskless replication (#7679) 2020-08-19 10:52:53 +03:00
guybe7
64cceb12ad PERSIST should signalModifiedKey (Like EXPIRE does) (#7671) 2020-08-18 19:07:59 +03:00
Oran Agra
1b5cc94836 OOM Crash log include size of allocation attempt. (#7670)
Since users often post just the crash log in github issues, the log
print that's above it is missing.
No reason not to include the size in the panic message itself.
2020-08-18 09:53:59 +03:00
Wen Hui
3962f80bd7 edit auth failed message (#7648)
Edit auth failed message include user disabled case in hello command
2020-08-18 08:59:24 +03:00
Wen Hui
b7236f0002 [module] using predefined REDISMODULE_NO_EXPIRE in RM_GetExpire (#7669)
It was already defined in the API header and the documentation, but not used by the implementation.
2020-08-18 08:50:03 +03:00
Oran Agra
05a4af3464 Trim trailing spaces in error replies coming from rejectCommand (#7668)
fe8d6fe749 added rejectCommand which takes an robj reply and passes it
through addReplyErrorSafe to addReplyErrorLength.
The robj contains newline at it's end, but addReplyErrorSafe converts it
to spaces, and passes it to addReplyErrorLength which adds the protocol
newlines.

The result was that most error replies (like OOM) had extra two trailing
spaces in them.
2020-08-18 08:28:43 +03:00
Yossi Gottlieb
2ec11f941a Module API: fix missing RM_CLIENTINFO_FLAG_SSL. (#7666)
The `REDISMODULE_CLIENTINFO_FLAG_SSL` flag was already a part of the `RedisModuleClientInfo` structure but was not implemented.
2020-08-17 17:46:54 +03:00
Yossi Gottlieb
24efd22e89 TLS: relax verification on CONFIG SET. (#7665)
Avoid re-configuring (and validating) SSL/TLS configuration on `CONFIG
SET` when TLS is not actively enabled for incoming connections, cluster
bus or replication.

This fixes failures when tests run without `--tls` on binaries that were
built with TLS support.

An additional benefit is that it's now possible to perform a multi-step
configuration process while TLS is disabled. The new configuration will
be verified and applied only when TLS is effectively enabled.
2020-08-17 17:36:50 +03:00
John Sully
dcf607622c Merge branch 'keydbpro' into PRO_RELEASE_6
Former-commit-id: ed98be0ba81ffdc501847ea0d2486f5f01391319
2020-08-17 02:26:51 +00:00
John Sully
93a67abe52 Fix ARM build of RocksDB
Former-commit-id: 7301d6d5e76b65115e2610a5fc5afee443782d7c
2020-08-17 02:25:45 +00:00
John Sully
e8e7061a73 Fast cleanup of snapshots without leaving them forever
Former-commit-id: fdd83b2b49244ed2988b080892ee5cffe9fd2684
2020-08-17 00:33:37 +00:00
John Sully
c0586b3aed Allow garbage collection of generic data
Former-commit-id: feadb7fb1845027422bcfca43dbcb6097409b8dc
2020-08-17 00:32:48 +00:00
Oran Agra
40ef9556c3 Merge pull request #7661 from michael-grunder/hiredis-unique-sds-symbols
This resolves an issue with sentinel that was created when hiredis was recently updated.
this was due to sds symbol names clashing, since hiredis now includes different implementation
of sdsrange than the one in redis.

The state of things is that redis-benchamrk and redis-cli include only hiredis sds implementation.
redis-cli even operates (calls sdscatlen) on sds that's allocated by hiredis.

Sentinel however has both implementations of the sds library in it (now each with it's own unique
symbols).
2020-08-16 08:54:36 +03:00
John Sully
00934cd341 Add snapshot depth to info string
Former-commit-id: 66aef678e9d6d7ab6c28622d54ada0ea8cdd2a99
2020-08-16 00:27:10 +00:00
John Sully
4a356ac13f Don't try and consolidate snapshots with a depth of 1
Former-commit-id: 26c298bd9bc4e2c6981de5c20284120ea54580c3
2020-08-16 00:26:05 +00:00
John Sully
3093d9e117 Only save MVCC tstamps if its an active replica
Former-commit-id: 5aded7dbad055308dc3932f7797fc71684b60966
2020-08-16 00:24:18 +00:00
John Sully
cf4e74006f Don't free snapshot objects in a critical path (under the AE lock)
Former-commit-id: d0da3d3cb74334cc8a2d14f4bdaef7935181700a
2020-08-16 00:13:19 +00:00
John Sully
311b286d41 BUG: Cannot have an EMBSTR bigger than 255
Former-commit-id: e2d6e2d9d585cb7a73f469a8580f9cb0ec71a429
2020-08-15 23:30:22 +00:00
John Sully
82989540c4 remove unnecessary allocation in rdb save
Former-commit-id: dfa76f04fd862e0cc7b4d3284b3922e3a5e5549d
2020-08-15 23:27:32 +00:00
John Sully
ce69a765f8 Remove unnecessary key comparisons in perf critical snapshot paths
Former-commit-id: 40f8a8d102fdca9443399ef03a47df609b146d58
2020-08-15 23:25:58 +00:00
John Sully
1055543817 dictMerge perf improvements
Former-commit-id: 48401ec369c5693689ef658cca518dc94ab1402e
2020-08-15 23:14:29 +00:00
John Sully
eb572c57e2 Rehash efficiency
Former-commit-id: fab383156626ec683881101c22eb2f6c2cea4c5d
2020-08-15 23:05:56 +00:00
John Sully
6c6ca43eef Prevent unnecessary copy when overwriting a value from a snapshot
Former-commit-id: 654a7bc6ea82f4ac45a1c1a25c794e1c27c0d902
2020-08-15 22:59:01 +00:00
John Sully
c804b28446 Merge branch 'unstable' into keydbpro
Former-commit-id: ae482585f0dc470efd73833f74111c2f87a172c5
2020-08-15 21:29:00 +00:00
michael-grunder
3db28bfe68 Use Hiredis' sdscompat.h to map sds* calls to hi_sds* 2020-08-15 13:13:23 -07:00
michael-grunder
82ae9ff451 Merge commit 'bffbbeaa9a1a6b8e81384297272cb0631502e8fd' into hiredis-unique-sds-symbols 2020-08-15 12:24:31 -07:00
michael-grunder
bffbbeaa9a Squashed 'deps/hiredis/' changes from d5b4c69b7..00272d669
00272d669 Rename sds calls so they don't conflict in Redis.

git-subtree-dir: deps/hiredis
git-subtree-split: 00272d669b11e96b8311d9bfe167c117f8887dd6
2020-08-15 12:24:31 -07:00
John Sully
6b8e979434 Prehash the tombstone for cleanup
Former-commit-id: c9d97a7c7448fc769486175bea1648589487c87c
2020-08-14 16:05:39 +00:00
Nathan Scott
9d4736b044 Annotate module API functions in redismodule.h for use with -fno-common (#6900)
In order to keep the redismodule.h self-contained but still usable with
gcc v10 and later, annotate each API function tentative definition with
the __common__ attribute.  This avoids the 'multiple definition' errors
modules will otherwise see for all API functions at link time.

Further details at gcc.gnu.org/gcc-10/porting_to.html

Turn the existing __attribute__ ((unused)), ((__common__)) and ((print))
annotations into conditional macros for any compilers not accepting this
syntax.  These macros only expand to API annotations under gcc.

Provide a pre- and post- macro for every API function, so that they can
be defined differently by the file that includes redismodule.h.

Removing REDISMODULE_API_FUNC in the interest of keeping the function
declarations readable.

Co-authored-by: Yossi Gottlieb <yossigo@gmail.com>
Co-authored-by: Oran Agra <oran@redislabs.com>
2020-08-14 14:45:34 +03:00
caozb
6ad1896ab3 wait command optimization (#7333)
Client that issued WAIT last will most likely have the highest replication offset, so imagine a probably common case where all clients are waiting for the same number of replicas. we prefer the loop to start from the last client (one waiting for the highest offset), so that the optimization in the function will call replicationCountAcksByOffset for each client until it found a good one, and stop calling it for the rest of the clients.
the way the loop was implemented would mean that in such case it is likely to call replicationCountAcksByOffset for all clients.

Note: the change from > to >= is not directly related to the above.

Co-authored-by: 曹正斌 <caozb@jiedaibao.com>
2020-08-13 14:30:51 +03:00
RemRain
8736ac973a Set the initial seed for random() (#5679) 2020-08-12 13:15:58 -07:00
Yossi Gottlieb
70c823a64e Add oom-score-adj configuration option to control Linux OOM killer. (#1690)
Add Linux kernel OOM killer control option.

This adds the ability to control the Linux OOM killer oom_score_adj
parameter for all Redis processes, depending on the process role (i.e.
master, replica, background child).

A oom-score-adj global boolean flag control this feature. In addition,
specific values can be configured using oom-score-adj-values if
additional tuning is required.
2020-08-12 17:58:56 +03:00
HarveyLiu
26bc18af61 fix misleading typo hasActiveChildProcess doc comment (#7588)
and a misspell in rax.c
2020-08-12 10:23:55 +03:00
Madelyn Olson
46ce35c55e Fixed timer warning (#5953) 2020-08-12 11:16:41 +08:00
Madelyn Olson
fed532b754 Replace usage of wrongtypeerr with helper (#7633)
* Replace usage of wrongtypeerr with helper
2020-08-11 20:04:54 -07:00
Mota
e637e1f02b Test:Fix invalid cases in hash.tcl and dump.tcl (#4611) 2020-08-12 10:25:24 +08:00
Mota
75fe12dbda Adds redis-cli and redis-benchmark dependencies for make test target
Obsoletes the need to run `make` before `make test`.
2020-08-11 22:01:15 +03:00