12492 Commits

Author SHA1 Message Date
John Sully
993359e1bb Fix race in db iterators in scan
Former-commit-id: a6444870660c0d3f52cd7b1dc0b80223f0d58e70
2020-07-09 22:29:27 +00:00
John Sully
bce5cb098b Do actual filtering on the async thread
Former-commit-id: 64b17139b7b5ff55bb52f27dc6f91cb81adde031
2020-07-09 21:58:06 +00:00
John Sully
29e4973d7b Initial scan implementation (not tested)
Former-commit-id: 0f3911d56e6878d91b6e3cacc7d637934ffe6099
2020-07-09 12:57:35 -04:00
John Sully
ffc3aefc41 Latency fixes
Former-commit-id: 6857c4c2085d3b0902d7b2ece66b3fe8828dd805
2020-07-01 21:59:38 -04:00
antirez
7bb0342201 Update comment to clarify change in #7398. 2020-06-25 12:58:21 +02:00
Salvatore Sanfilippo
339fb2de8d Merge pull request #7398 from caiyuxinggg/work
cluster.c remove "if (nodeIsMaster(myself))" judgement before clusterSendFail in markNodeAsFailingIfNeeded, avoiding slave failover requires twice vote requests
2020-06-25 12:56:26 +02:00
Salvatore Sanfilippo
03ab78d115 Merge pull request #7426 from dnielsen/patch-1
updated copyright year to 2020
2020-06-25 12:36:37 +02:00
Muhammad Zahalqa
f024fe30f4 replace memcpy with cctor on objects
Former-commit-id: 52897f8f9a882bcdbd1e8ede6bdf24e7435f5ce8
2020-06-24 23:20:59 -04:00
Muhammad Zahalqa
ec28161271 unique lock used with cv.wait already owns mutex
Former-commit-id: 8bc70e9840222d11507af30a6db27210b16650a2
2020-06-24 23:18:53 -04:00
Muhammad Zahalqa
63bb6cb419 Fix bug in condition_variabe.wait usage
consition variable wait should always be called with a locked mutex.
This code break the pre-conditions of using wait.


Former-commit-id: 5b81303f04580010f043e518a109b43971a63054
2020-06-24 23:18:53 -04:00
Muhammad Zahalqa
e25ec37484 fixes for robj_sharedptr
1. fix cases where null pointer might be accessed
2. make assignmnet op safe
3. make operator bool explicit (safe bool idiom)
4. make comparison operators symetric

fix robj_sharedptr use in rdb.cpp


Former-commit-id: ede524c0647c0875f1071978f26ff785c8d1183e
2020-06-24 23:18:27 -04:00
Muhammad Zahalqa
c4cd846388 aeCommand objects no need for memset and missing init of some memebers.
Former-commit-id: aeed09c34620e465b57f1d83553d259c4f6d892f
2020-06-24 23:15:29 -04:00
antirez
e2f5953c73 Merge branch 'unstable' of github.com:/antirez/redis into unstable 2020-06-24 09:09:59 +02:00
antirez
5b16c2f174 LPOS: option FIRST renamed RANK. 2020-06-24 09:09:43 +02:00
benschermel
cd8b82db55 Merge branch 'keydbpro' into PRO_RELEASE_6
Former-commit-id: 6e9ec7ae3ca49a6054f012e567eb74307227c330
2020-06-24 00:34:35 -04:00
benschermel
9c2769804e update deb changelog
Former-commit-id: 85fe16238fdb692067f77f631b0589366823ba34
2020-06-24 00:32:10 -04:00
John Sully
638926621a Merge PKG files from unstable
No code change


Former-commit-id: 42c4d85d72af99322b21faab74ae23556b2054f0
2020-06-24 00:16:21 -04:00
Dave Nielsen
e54ea02c43 updated copyright year
Changed "2015" to "2020"
2020-06-23 09:51:12 -07:00
Salvatore Sanfilippo
54a422f6e0 Merge pull request #7390 from oranagra/exec_fails_abort
EXEC always fails with EXECABORT and multi-state is cleared
2020-06-23 13:12:52 +02:00
Oran Agra
fe8d6fe749 EXEC always fails with EXECABORT and multi-state is cleared
In order to support the use of multi-exec in pipeline, it is important that
MULTI and EXEC are never rejected and it is easy for the client to know if the
connection is still in multi state.

It was easy to make sure MULTI and DISCARD never fail (done by previous
commits) since these only change the client state and don't do any actual
change in the server, but EXEC is a different story.

Since in the past, it was possible for clients to handle some EXEC errors and
retry the EXEC, we now can't affort to return any error on EXEC other than
EXECABORT, which now carries with it the real reason for the abort too.

Other fixes in this commit:
- Some checks that where performed at the time of queuing need to be re-
  validated when EXEC runs, for instance if the transaction contains writes
  commands, it needs to be aborted. there was one check that was already done
  in execCommand (-READONLY), but other checks where missing: -OOM, -MISCONF,
  -NOREPLICAS, -MASTERDOWN
- When a command is rejected by processCommand it was rejected with addReply,
  which was not recognized as an error in case the bad command came from the
  master. this will enable to count or MONITOR these errors in the future.
- make it easier for tests to create additional (non deferred) clients.
- add tests for the fixes of this commit.
2020-06-23 12:01:33 +03:00
antirez
79e15d1686 Include cluster.h for getClusterConnectionsCount(). 2020-06-22 11:44:11 +02:00
antirez
448a8bb0ed Fix BITFIELD i64 type handling, see #7417. 2020-06-22 11:41:19 +02:00
antirez
c10a7a040f Clarify maxclients and cluster in conf. Remove myself too. 2020-06-22 11:21:21 +02:00
Salvatore Sanfilippo
4d45ec35b2 Merge pull request #7421 from hwware/sentinel_leak_fix
fix memory leak in sentinel connection sharing
2020-06-22 11:06:50 +02:00
hwware
bdac36b70e fix memory leak in sentinel connection sharing 2020-06-21 23:04:28 -04:00
Salvatore Sanfilippo
bf12f73224 Merge pull request #7402 from chenhui0212/unstable
Fix comments in listpack.c
2020-06-18 11:29:06 +02:00
chenhui0212
44235ac718 Fix comments in function raxLowWalk of listpack.c 2020-06-18 17:28:26 +08:00
Salvatore Sanfilippo
66657b0001 Merge pull request #7411 from tporadowski/inconsitent-sentinel-nosave
Inconsistent RDB saving when in Sentinel mode
2020-06-18 11:26:53 +02:00
Tomasz Poradowski
6782243054 ensure SHUTDOWN_NOSAVE in Sentinel mode
- enforcing of SHUTDOWN_NOSAVE flag in one place to make it consitent
  when running in Sentinel mode
2020-06-17 22:22:49 +02:00
Ben Schermel
4d6b7c278b bump version after merge
Former-commit-id: 59279d36e9337ca89518c4d1edef6c26911ea190
2020-06-16 14:17:20 -04:00
Ben Schermel
e0d2a1cff6 Merge branch 'keydbpro' into PRO_RELEASE_6
Former-commit-id: cf7ed118553af55a27c3a57c121bc26e18e0ab7e
2020-06-16 14:16:03 -04:00
Ben Schermel
38f5ed20a0 Merge branch 'PRO_RELEASE_6' of https://github.com/JohnSully/KeyDB-Pro into PRO_RELEASE_6
Former-commit-id: ff7ec3ec39d692ed59c945ace5c9a4ec140f5e52
2020-06-16 14:15:51 -04:00
chenhui0212
6b82471098 fix comments in listpack.c 2020-06-16 17:50:38 +08:00
antirez
fc08cafdb0 Use cluster connections too, to limit maxclients.
See #7401.
2020-06-16 11:45:11 +02:00
antirez
714c2be383 Merge branch 'unstable' of github.com:/antirez/redis into unstable 2020-06-16 11:11:01 +02:00
antirez
ae770c3034 Tracking: fix enableBcastTrackingForPrefix() invalid sdslen() call.
Related to #7387.
2020-06-16 11:09:48 +02:00
John Sully
e81a4a159a Fix crash in async delete when storage provider is not set and key doesn't actually exist
Former-commit-id: 4a4828390118ab52851d8bf229c4ae853100053c
2020-06-15 19:47:06 -04:00
Salvatore Sanfilippo
96ca6e9f5d Merge pull request #7396 from MeirShpilraien/fix_dict_scan_module_api
Fix RM_ScanKey module api not to return int encoded strings
2020-06-15 10:43:33 +02:00
root
009a2d443a cluster.c remove if of clusterSendFail in markNodeAsFailingIfNeeded 2020-06-15 10:18:14 +08:00
meir@redislabs.com
e37c16e425 Fix RM_ScanKey module api not to return int encoded strings
The scan key module API provides the scan callback with the current
field name and value (if it exists). Those arguments are RedisModuleString*
which means it supposes to point to robj which is encoded as a string.
Using createStringObjectFromLongLong function might return robj that
points to an integer and so break a module that tries for example to
use RedisModule_StringPtrLen on the given field/value.

The PR introduces a fix that uses the createObject function and sdsfromlonglong function.
Using those function promise that the field and value pass to the to the
scan callback will be Strings.

The PR also changes the Scan test module to use RedisModule_StringPtrLen
to catch the issue. without this, the issue is hidden because
RedisModule_ReplyWithString knows to handle integer encoding of the
given robj (RedisModuleString).

The PR also introduces a new test to verify the issue is solved.
2020-06-14 11:20:15 +03:00
antirez
00e1f87a08 Fix LCS object type checking. Related to #7379. 2020-06-12 12:43:40 +02:00
Salvatore Sanfilippo
c35a814c4f Merge pull request #7375 from hwware/lcs_crash_fix
Fix Server Crash in LCS Command
2020-06-12 12:31:15 +02:00
antirez
d05cc51e4e help.h updated. 2020-06-12 12:16:24 +02:00
antirez
f66fc18807 help.h updated. 2020-06-12 12:16:19 +02:00
antirez
3dc6657ea8 LPOS: tests + crash fix. 2020-06-12 12:08:06 +02:00
antirez
010dd2b320 LPOS: update to latest proposal.
See https://gist.github.com/antirez/3591c5096bc79cad8b5a992e08304f48
2020-06-12 12:08:06 +02:00
antirez
00e400ed66 LPOS: implement the final design. 2020-06-12 12:08:06 +02:00
Paul Spooren
795c807fed LRANK: Add command (the command will be renamed LPOS).
The `LRANK` command returns the index (position) of a given element
within a list. Using the `direction` argument it is possible to specify
going from head to tail (acending, 1) or from tail to head (decending,
-1). Only the first found index is returend. The complexity is O(N).

When using lists as a queue it can be of interest at what position a
given element is, for instance to monitor a job processing through a
work queue. This came up within the Python `rq` project which is based
on Redis[0].

[0]: https://github.com/rq/rq/issues/1197

Signed-off-by: Paul Spooren <mail@aparcar.org>
2020-06-12 12:08:06 +02:00
hwware
8ab655bd7b fix memory leak 2020-06-11 09:56:52 -04:00
antirez
b7db223637 LPOS: tests + crash fix. 2020-06-11 12:39:06 +02:00