gvsafronov/futriix
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
10,837 Commits 1 Branch 0 Tags
Commit Graph

7834 Commits

Author SHA1 Message Date
John Sully
9204f7b629 Fix race when we free an item before we finish writing to disk making it temporarily unavailable 
Former-commit-id: 6139cfee295261d0210f2af6217c2a23eeaf52b8
 2020-06-04 22:18:03 -04:00
John Sully
e522540e2f Add latency marker for process changes phase for FLASH 
Former-commit-id: 61e20b025d45b0af8ed60aab96246b30e4d0ed3d
 2020-06-04 01:29:17 -04:00
John Sully
bec9bb1e0f Latency doctor updated for KeyDB Pro 
Former-commit-id: a2690fca4a74e92eb44d5e38f886b08363a6d7b0
 2020-06-04 01:28:52 -04:00
John Sully
689c1354be maxmem samples way too high 
Former-commit-id: 599cac856d4ec77d59c5f5b4fae5b9f194c202ef
 2020-06-04 01:28:34 -04:00
John Sully
8888498bfd Make snapshot completion faster and add latency monitor 
Former-commit-id: 8063be6ee70a652c22c3263dccf318366e208891
 2020-06-04 01:07:14 -04:00
John Sully
cb4bb9d7eb Consolidate snapshots should not consider keys in bulk storage 
Former-commit-id: e02a8f503d45daf63150c60a8a8c3751b9b8581e
 2020-06-04 00:26:51 -04:00
John Sully
e7586ddca1 Disambiguate temp rdb files so its less likely two servers with the same working dir will collide 
Former-commit-id: 8fbc5a63c54589bb42aa1b1b53c56f399211605c
 2020-06-04 00:25:37 -04:00
John Sully
f50ed30ad2 Only set tombstone if there is something to mask 
Former-commit-id: ea092693f53ff0803c44a851219993a933166cc3
 2020-06-04 00:23:37 -04:00
John Sully
28397f140a Extra debug checks in dictmerge 
Former-commit-id: 839b2a151e6bad67017de87d7a637359f6ae63d1
 2020-06-04 00:22:59 -04:00
John Sully
b39f917335 Clear tombstone dict when clearing database we don't need it 
Former-commit-id: a8e1ef075eacfc0b767302f2514ca6f0342982bc
 2020-06-04 00:17:19 -04:00
John Sully
4cf14b6e49 Fix crash where tombstone is evicted 
Former-commit-id: 1f6ff7cc0fc0173610d77a69076eeaf478c3b701
 2020-06-03 23:30:35 -04:00
John Sully
cebc1b7e48 Add DB commit latency samples 
Former-commit-id: 557ea8d285045496105e3330f0007e02aa4359d7
 2020-06-03 22:06:05 -04:00
John Sully
589448bdf0 Fix crash when overwriting key stored in FLASH but not in ram during a BGSAVE while clustering is enabled 
Former-commit-id: 32967b1d671b151bed976e3d13ce26c7e1496c82
 2020-06-02 23:59:24 -04:00
John Sully
ffbd43fbfa Bump maxmemory samples by a lot 
Former-commit-id: ae46d64b2e6b16f81df6b4e12ddc9fed63a9785b
 2020-06-02 17:22:45 -04:00
John Sully
ad174ab352 Merge error, server always flushes on eviction 
Former-commit-id: 9b3b0c9a40ab5dcdac9620f0fbad84887cc9c832
 2020-06-02 17:21:41 -04:00
John Sully
64adf015b1 Undecorated new is OK to use 
Former-commit-id: 5b885bb1649805f6a2edb8d28edd1447bb6c4843
 2020-06-01 23:36:01 -04:00
John Sully
8bdca52222 Merge branch 'keydbpro' into PRO_RELEASE_6 
Former-commit-id: a901505e562c0dc341fa939cce502da25642f7ba
 2020-06-01 17:42:32 -04:00
John Sully
df3f1e8d8e Merge branch 'unstable' into keydbpro 
Former-commit-id: 08a36155e3db9918048e87c3d691b7317787c9ab
 2020-06-01 17:41:37 -04:00
John Sully
c799f6855b Bump version 
Former-commit-id: 8dbea2a050dd9e741018230fedeeda46780a7a31
 2020-06-01 17:01:50 -04:00
John Sully
b2a885c98f Remove debug logs that shouldn't have been checked in 
Former-commit-id: 31f58311e3de7441d81dd37bd4396be3b64efec5
 2020-06-01 16:34:05 -04:00
John Sully
9e87395c34 Fix for issue #187 we need to properly handle the case where a key with a subkey expirey itself expires during load 
Former-commit-id: e6a9a6b428b91b6108df24ae6285ea9b582b7b23
 2020-06-01 15:33:19 -04:00
John Sully
0a9a32e5d7 Fix module multithreaded test failures 
Former-commit-id: 1ef35cf466ea944c56974b3795d7d6b5e89f5a3d
 2020-05-31 23:46:12 -04:00
John Sully
08fca5ef31 sendfile has high latency in some scenarios, don't use it 
Former-commit-id: 1eb0e3c1c604e71c54423f1d11b8c709c847a516
 2020-05-31 23:22:25 -04:00
John Sully
b30fa046f4 we must always respect the output of sdscat, also change the string so its not a substring of other config params 
Former-commit-id: e50313b20718bc8df0a53c11b0960e4bcb2177d4
 2020-05-31 23:22:10 -04:00
John Sully
4b317392be Don't start multimaster tests until all nodes are connected 
Former-commit-id: 202b97eff76501e736a2f0969607e3297e9703a4
 2020-05-31 22:50:30 -04:00
John Sully
7eb0563241 Active replicas referencing eachother should connect one at a time 
Former-commit-id: c0c033a0c175eebdf2173e6e4e59e792d2fe4285
 2020-05-31 16:03:50 -04:00
John Sully
891977d9b7 Fix race in futex_sleep 
Former-commit-id: 73300c57005a49bafbfc44db9c40ba7d1d4eedce
 2020-05-31 16:03:50 -04:00
John Sully
d344d9b0b9 Auto tune lock for high CPU tension scenarios 
Former-commit-id: 8edbae2e04538f82a146a6c2b459a6dfcacf99b2
 2020-05-30 18:22:27 -04:00
John Sully
baadab90e7 We only want one master connection in progress at a time, ensure that if the connection fails we try a different master 
Former-commit-id: 8a0441c14475dc54616337270e092068acaa274a
 2020-05-30 15:13:28 -04:00
John Sully
f37ee5bd4d TLS requires we explicitly marshal connected sockets when using them on a different thread 
Former-commit-id: 341a8caef959883ca6fc71f81f8a3fed747b2341
 2020-05-30 01:14:53 -04:00
Ben Schermel
be6f1df429 centos build fix 
Former-commit-id: f17c492a633ea2be99567dfe2f8a59c0b0136d86
 2020-05-29 22:23:48 -04:00
John Sully
5316f656d9 sendFile blocks too long for use with active replication 
Former-commit-id: aad6a7ce159a3679633020dc407a2068129bbd49
 2020-05-29 20:27:45 -04:00
John Sully
68bf5a04a0 Unify beforeSleep handling 
Former-commit-id: 1cb48c7bf6a7e91e728a677902a7bfc64fe80dd6
 2020-05-29 17:44:55 -04:00
John Sully
f18ba66242 Cluster crash 
Former-commit-id: f25c405ad2a8004b79a816072cb011c1dc2407d5
 2020-05-29 03:40:52 -04:00
John Sully
ed2e0e66f6 Merge tag '6.0.4' into unstable 
Redis 6.0.4.


Former-commit-id: 9c31ac7925edba187e527f506e5e992946bd38a6
 2020-05-29 00:57:07 -04:00
John Sully
71fbeb65b6 fix TLS test failure 
Former-commit-id: 57ca6facc3038e005656912b6378cb7fbe04f55a
 2020-05-28 23:16:27 -04:00
antirez
c512f64428 Redis 6.0.4. 2020-05-28 12:18:38 +02:00
antirez
84117d13b7 Replication: showLatestBacklog() refactored out. 2020-05-28 10:09:51 +02:00
antirez
14d99c183f Drop useless line from replicationCacheMaster(). 2020-05-28 10:09:51 +02:00
antirez
911c579b68 Remove the meaningful offset feature. 
After a closer look, the Redis core devleopers all believe that this was
too fragile, caused many bugs that we didn't expect and that were very
hard to track. Better to find an alternative solution that is simpler.
 2020-05-28 10:09:51 +02:00
antirez
7e55485b21 Set a protocol error if master use the inline protocol. 
We want to react a bit more aggressively if we sense that the master is
sending us some corrupted stream. By setting the protocol error we both
ensure that the replica will disconnect, and avoid caching the master so
that a full SYNC will be required. This is protective against
replication bugs.
 2020-05-28 10:09:51 +02:00
Oran Agra
0705a29959 avoid using sendfile if tls-replication is enabled 
this obviously broke the tests, but went unnoticed so far since tls
wasn't often tested.
 2020-05-28 10:09:51 +02:00
antirez
fee0c76304 Replication: log backlog creation event. 2020-05-28 10:09:51 +02:00
John Sully
8ff703992c Merge branch 'keydbpro' into PRO_RELEASE_6 
Former-commit-id: f98c1999f95ec59f7201056781cb89dc465a1461
 2020-05-26 01:38:52 -04:00
John Sully
ef32f81fd9 Complete TLS merge 
Former-commit-id: 856d23023259d59ce766e9cd00526de0e7d06c2f
 2020-05-26 01:38:29 -04:00
John Sully
f617eafbe1 Merge branch 'unstable' into keydbpro 
Former-commit-id: b0636273806ce323627ce31c1ad7f86ceb39696d
 2020-05-26 01:34:19 -04:00
John Sully
ef4284c439 Merge branch 'unstable' of https://github.com/JohnSully/KeyDB into unstable 
Former-commit-id: 1449c8b9fe647288da7de0ff10cc68ada4be8f92
 2020-05-26 01:29:08 -04:00
John Sully
5b44b62854 Fix client tracking, also ensure tracking tests are enabled 
Former-commit-id: 1938af27f50f9686dc98b4839fb439cc03b4a250
 2020-05-26 01:28:52 -04:00
antirez
7a32a8485e Clarify what is happening in PR #7320. 2020-05-25 12:08:01 +02:00
zhaozhao.zz
d089cc8963 PSYNC2: second_replid_offset should be real meaningful offset 
After adjustMeaningfulReplOffset(), all the other related variable
should be updated, including server.second_replid_offset.

Or the old version redis like 5.0 may receive wrong data from
replication stream, cause redis 5.0 can sync with redis 6.0,
but doesn't know meaningful offset.
 2020-05-25 12:08:01 +02:00