7489 Commits

Author SHA1 Message Date
antirez
643ee6e38c When replica kills a pending RDB save during SYNC, log it.
This logs what happens in the context of the fix in PR #5367.
2018-11-05 17:12:37 +01:00
Andrey Bugaevskiy
8b609c9986 Move child termination to readSyncBulkPayload 2018-11-05 17:12:37 +01:00
Andrey Bugaevskiy
2710260594 Prevent RDB autosave from overwriting full resync results
During the full database resync we may still have unsaved changes
on the receiving side. This causes a race condition between
synced data rename/load and the rename of rdbSave tempfile.
2018-11-05 17:12:37 +01:00
antirez
a677923d7c asyncCloseClientOnOutputBufferLimitReached(): don't free fake clients.
Fake clients are used in special situations and are not linked to the
normal clients list, freeing them will always result in Redis crashing
in one way or the other.

It's not common to send replies to fake clients, but we have one usage
in the modules API. When a client is blocked, we associate to the
blocked client object (that is safe to manipulate in a thread), a fake
client that accumulates replies. So because of this bug there was
the problem described in issue #5443.

The fix was verified to work with the provided example module. To write
a regression is very hard and unlikely to be triggered in the future.
2018-11-05 17:12:37 +01:00
David Carlier
427e440a58 needs it for the global 2018-11-05 17:12:37 +01:00
David Carlier
28f9ca4e1d Fix non Linux build.
timezone global is a linux-ism whereas it is a function under BSD.
Here a helper to get the timezone value in a more portable manner.
2018-11-05 17:12:37 +01:00
zhaozhao.zz
4bf9efe20f MULTI: OOM err if cannot free enough memory in MULTI/EXEC context 2018-11-05 17:12:37 +01:00
antirez
4fbd7a39f8 Add command fingerprint comment for XSETID. 2018-11-05 17:12:37 +01:00
Itamar Haber
2480db53f1 Plugs a potential underflow 2018-11-05 17:12:37 +01:00
Itamar Haber
e5e4d2ef08 Corrects inline documentation of syntax 2018-11-05 17:12:37 +01:00
zhaozhao.zz
713800d20a if we read a expired key, misses++ 2018-11-05 17:12:37 +01:00
antirez
e79ee263cf Fix XRANGE COUNT option for value of 0. 2018-11-05 17:12:37 +01:00
antirez
505cc70ff8 Fix typo in streamReplyWithRange() top comment. 2018-11-05 17:12:37 +01:00
Damien Tournoud
3c36561d27 Overhead is the allocated size of the AOF buffer, not its length 2018-11-05 17:12:37 +01:00
antirez
3761582ffc Simplify part of the #5470 patch. 2018-11-05 17:12:37 +01:00
zhaozhao.zz
edc47a3ad2 do not delete expired keys in KEYS command 2018-11-05 17:12:37 +01:00
antirez
9872af6d52 Use guide comments to make changes in #5462 more obvious. 2018-10-22 17:45:39 +02:00
youjiali1995
3f399c3bff migrate: fix mismatch of RESTORE reply when some keys have expired. 2018-10-22 17:45:39 +02:00
hujie
eaaff6211b fix typo in config.c 2018-10-22 17:45:39 +02:00
hujiecs
43ebb7ee01 several typos fixed, optimize MSETNX to avoid unnecessary loop 2018-10-22 17:45:31 +02:00
antirez
de8fdaacfc Remove useless complexity from MSET implementation. 2018-10-22 17:44:26 +02:00
antirez
dc8f111251 Fix again stack generation on the Raspberry Pi.
The fix was removed by c8ca71d40 attempting to fix the stack generation
on ARM64, without testing if it would still work on ARM32.
Now it should work both sides.
2018-10-19 10:40:42 +02:00
antirez
83a6e81db3 Get rid of the word slave in the release note of Redis 5. 2018-10-17 18:11:49 +02:00
antirez
882ca6962f Redis 5.0.0. 5.0.0 2018-10-17 17:31:39 +02:00
antirez
1eb0994ecc Streams: use bulk replies instead of status replies.
They play better with Lua scripting, otherwise Lua will see status
replies as "ok" = "string" which is very odd, and actually as @oranagra
reasoned in issue #5456 in the rest of the Redis code base there was no
such concern as saving a few bytes when the protocol is emitted.
2018-10-17 17:20:05 +02:00
antirez
bcc0916d08 Fix conditional in XGROUP. 2018-10-17 13:00:55 +02:00
antirez
1b2f23f3c4 Update help.h for redis-cli. 2018-10-17 12:57:15 +02:00
antirez
de0ae56c84 Tests for XGROUP CREATE MKSTREAM. 2018-10-17 12:12:04 +02:00
antirez
56c3dfa195 Fix XGROUP CREATE MKSTREAM handling of . 2018-10-17 12:11:48 +02:00
antirez
2687f2283c Process MKSTREAM option of XGROUP CREATE at a later time.
This avoids issues with having to replicate a command that produced
errors.
2018-10-17 12:11:48 +02:00
zhaozhao.zz
cfbaf8f1f3 Scripting & Streams: some commands need right flags
xadd with id * generates random stream id

xadd & xtrim with approximate maxlen count may
trim stream randomly

xinfo may get random radix-tree-keys/nodes

xpending may get random idletime

xclaim: master and slave may have different
idletime in stream
2018-10-17 12:11:48 +02:00
antirez
4e4099b95d XGROUP CREATE: MKSTREAM option for automatic stream creation. 2018-10-17 11:27:09 +02:00
zhaozhao.zz
6dd4d864a9 Streams: Tests modified XSTREAM -> XSETID 2018-10-17 11:02:34 +02:00
zhaozhao.zz
3aff0e8cb5 Streams: rewrite empty streams with certain lastid 2018-10-17 11:02:34 +02:00
antirez
880b563e2d Tests modified to use XADD MAXLEN 0 + XSETID.
See #5426.
2018-10-17 11:02:34 +02:00
antirez
83c8783598 Streams: rewrite empty streams with XADD MAXLEN 0. Use XSETID.
Related to #5426.
2018-10-17 11:02:34 +02:00
antirez
fd22e3acae XSETID: accept IDs based on last entry.
Related to #5426.
2018-10-17 11:02:34 +02:00
antirez
dfab3cba2a Streams: XSTREAM SETID -> XSETID.
Keep vanilla stream commands at toplevel, see #5426.
2018-10-17 11:02:34 +02:00
zhaozhao.zz
a3fb28edce Streams: rewrite id in XSTREAM CREATE * 2018-10-17 11:02:34 +02:00
zhaozhao.zz
f4b4db1387 Streams: add tests for aof rewrite 2018-10-17 11:02:34 +02:00
zhaozhao.zz
d22f1ef032 Stream & AOF: rewrite stream in correct way 2018-10-17 11:02:34 +02:00
zhaozhao.zz
6455274d1c Streams: add tests for XSTREAM command 2018-10-17 11:02:34 +02:00
zhaozhao.zz
0edbe953ea Streams: add a new command XTREAM
XSTREAM CREATE <key> <id or *> -- Create a new empty stream.
XSTREAM SETID <key> <id or $>  -- Set the current stream ID.
2018-10-17 11:02:34 +02:00
Hamid Alaei
9714bba266 fix timer context selected database 2018-10-15 13:05:42 +02:00
antirez
eb53f15a3e Make comment about nack->consumer test for minidle more obvious.
Related to #5437.
2018-10-15 13:05:35 +02:00
antirez
a77f836e0d Streams: use propagate_last_id itself as streamPropagateGroupID trigger.
Avoid storing the dirty value. See #5437.
2018-10-15 13:05:35 +02:00
antirez
0f0610eb01 Streams: better naming: lastid_updated -> propagate_last_id.
See #5437 but also I updated a previous usage of the same var name.
2018-10-15 13:05:35 +02:00
zhaozhao.zz
a745e4234d Streams: panic if streamID invalid after check, should not be possible. 2018-10-15 13:05:35 +02:00
zhaozhao.zz
9974be13a4 Streams: propagate lastid in XCLAIM when it has effect 2018-10-15 13:05:35 +02:00
zhaozhao.zz
69a628d0f0 Streams: XCLAIM ignore minidle if NACK is created by FORCE
Because the NACK->consumer is NULL, if idletime < minidle
the NACK does not belong to any consumer, then redis will crash
in XPENDING.
2018-10-15 13:05:35 +02:00