Commit Graph

7266 Commits

Author SHA1 Message Date
e8d35809c4 While the slave is busy, just accumulate master input.
Processing command from the master while the slave is in busy state is
not correct, however we cannot, also, just reply -BUSY to the
replication stream commands from the master. The correct solution is to
stop processing data from the master, but just accumulate the stream
into the buffers and resume the processing later.

Related to #5297.
2018-08-30 17:52:04 +02:00
4b68dbcbd7 Allow scripts to timeout even if from the master instance.
However the master scripts will be impossible to kill.

Related to #5297.
2018-08-30 11:46:42 +02:00
17b0d99d9f Allow scripts to timeout on slaves as well.
See reasoning in #5297.
2018-08-29 18:14:46 +02:00
476eea95da Merge pull request #4216 from lamby/did-not-received-typos
Correct "did not received" -> "did not receive" typos/grammar.
2018-08-29 16:18:11 +02:00
bd9259d12d Merge pull request #5282 from soloestoy/remove-duplicate-bind-in-sentinel.conf
remove duplicate bind in sentinel.conf
2018-08-29 16:13:42 +02:00
ed5cc77ce0 Merge pull request #5296 from soloestoy/command-script-flag2
Supplement to PR #4835, just take info/memory/command as random commands
2018-08-29 12:26:01 +02:00
7d39c149c4 Supplement to PR #4835, just take info/memory/command as random commands 2018-08-29 18:23:05 +08:00
0e21efdb1c Merge pull request #4835 from soloestoy/command-script-flag
some commands' flags should be set correctly, issue #4834
2018-08-29 12:13:50 +02:00
32844178ac some commands' flags should be set correctly, issue #4834 2018-08-29 18:07:01 +08:00
5b4bec9d33 Merge pull request #5265 from oranagra/stabilize_tests
Fix unstable tests on slow machines.
2018-08-27 13:19:31 +02:00
5941022b52 Document slave-ignore-maxmemory in redis.conf. 2018-08-27 12:34:29 +02:00
e245a2046a Make slave-ignore-maxmemory configurable. 2018-08-27 12:27:17 +02:00
067647a783 Introduce repl_slave_ignore_maxmemory flag internally.
Note: this breaks backward compatibility with Redis 4, since now slaves
by default are exact copies of masters and do not try to evict keys
independently.
2018-08-27 12:20:27 +02:00
abf52c7cf4 Better variable meaning in processCommand(). 2018-08-27 12:17:34 +02:00
66b5afdaa4 Re-apply rebased #2358. 2018-08-27 12:17:14 +02:00
c241f51607 Fix build errors caused by #2358. 2018-08-27 12:15:55 +02:00
e711e6f614 remove duplicate bind in sentinel.conf 2018-08-27 12:07:24 +08:00
19880ab851 Merge pull request #5248 from soloestoy/rewrite-brpoplpush
rewrite BRPOPLPUSH as RPOPLPUSH to propagate
2018-08-26 16:31:24 +02:00
80e1695652 Merge pull request #5244 from soloestoy/optimize-pipeline
pipeline: do not sdsrange querybuf unless all commands processed
2018-08-26 16:30:49 +02:00
132be8aed5 Correct "did not received" -> "did not receive" typos/grammar. 2018-08-26 14:45:39 +02:00
46d89a9abb Merge pull request #2992 from lamby/source-date-epoch
Use SOURCE_DATE_EPOCH over unreproducible uname + date calls.
2018-08-26 11:25:44 +02:00
590717782f Merge pull request #2358 from lamby/config-set-maxmemory-grammar
Tidy grammar in CONFIG SET maxmemory warning.
2018-08-26 11:23:41 +02:00
8d519a85f4 Merge pull request #2292 from lamby/sentinel-conf-defaults
Make some defaults explicit in the sentinel.conf for package maintainers
2018-08-26 11:19:51 +02:00
f63e81c202 Merge branch 'unstable' into config-set-maxmemory-grammar 2018-08-25 21:49:29 +02:00
f2ad89a314 networking: make setProtocolError simple and clear
Function setProtocolError just records proctocol error
details in server log, set client as CLIENT_CLOSE_AFTER_REPLY.
It doesn't care about querybuf sdsrange, because we
will do it after procotol parsing.
2018-08-23 12:21:28 +08:00
c8452ab005 Fix unstable tests on slow machines.
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
2018-08-21 11:46:07 +03:00
9a65f9cd3e block: format code 2018-08-14 20:59:32 +08:00
8a1219d93b block: rewrite BRPOPLPUSH as RPOPLPUSH to propagate 2018-08-14 20:58:58 +08:00
ef2a95c461 networking: just move qb_pos instead of sdsrange in processInlineBuffer 2018-08-14 14:50:37 +08:00
e623bd22ba networking: just return C_OK if multibulk processing saw a <= 0 length. 2018-08-14 13:55:30 +08:00
b89302c462 adjust qbuf to 26 in test case for client list 2018-08-14 00:57:22 +08:00
14c4ddb5a6 pipeline: do not sdsrange querybuf unless all commands processed
This is an optimization for processing pipeline, we discussed a
problem in issue #5229: clients may be paused if we apply `CLIENT
PAUSE` command, and then querybuf may grow too large, the cost of
memmove in sdsrange after parsing a completed command will be
horrible. The optimization is that parsing all commands in queyrbuf
, after that we can just call sdsrange only once.
2018-08-14 00:43:42 +08:00
39c70e728b Merge pull request #5146 from 0xtonyxia/fix-xclaim-id-parse
Streams: ID of xclaim command should start from the sixth argument.
2018-08-03 13:45:27 +02:00
7b5e7f3ed0 Merge pull request #5151 from shenlongxing/fix-stream
Fix stream command paras
2018-08-03 13:39:32 +02:00
fe56c67405 Fix stream command paras 2018-08-03 19:01:15 +08:00
0ce8323c0d Fix AOF comment to report the current behavior.
Realted to #5201.
2018-08-03 12:46:06 +02:00
3d56311f0c Test suite: add --loop option.
Very useful with --stop in order to catch heisenbugs.
2018-08-02 19:07:17 +02:00
26897c03e2 Test suite: new --stop option.
It pauses the test execution once the first failure is found.
2018-08-02 18:49:49 +02:00
d506334b67 Test: new sorted set skiplist order consistency.
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.
2018-08-02 14:15:53 +02:00
2f282aee0b Fix zslUpdateScore() edge case.
When the element new score is the same of prev/next node, the
lexicographical order kicks in, so we can safely update the node in
place only when the new score is strictly between the adjacent nodes
but never equal to one of them.

Technically speaking we could do extra checks to make sure that even if the
score is the same as one of the adjacent nodes, we can still update on
place, but this rarely happens, so probably not a good deal to make it
more complex.

Related to #5179.
2018-08-01 19:04:53 +02:00
29226a3919 More commenting of zslUpdateScore(). 2018-08-01 18:54:15 +02:00
12ff0c0d79 Explain what's the point of zslUpdateScore() in top comment. 2018-08-01 18:53:09 +02:00
e3e94ec35c Remove old commented zslUpdateScore() from source. 2018-08-01 18:50:49 +02:00
0b800c4332 Optimize zslUpdateScore() as asked in #5179. 2018-08-01 18:50:31 +02:00
6dd0d6f7dd zsetAdd() refactored adding zslUpdateScore(). 2018-08-01 18:27:56 +02:00
2f2987ffc5 Streams IDs parsing refactoring.
Related to #5184.
2018-07-31 18:08:52 +02:00
187fa78637 Merge pull request #5184 from 0xtonyxia/streams-id-opt
Streams: rearrange the usage of '-' and '+' IDs in stream commands.
2018-07-31 17:58:13 +02:00
7a30be1237 Minor improvements to PR #5187. 2018-07-31 17:30:12 +02:00
7b41701eb8 Merge pull request #5187 from oranagra/testsuite_improvements
Testsuite improvements
2018-07-31 17:20:09 +02:00
e245ed9a44 Cluster cron announce IP minor refactoring. 2018-07-31 17:13:03 +02:00