gvsafronov/futriix
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
12,001 Commits 1 Branch 0 Tags
Commit Graph

8734 Commits

Author SHA1 Message Date
Salvatore Sanfilippo
37b318c7e0 Merge pull request #7363 from xhebox/unstable 
return the correct proto version
 2020-06-08 12:54:15 +02:00
Salvatore Sanfilippo
5f1defa179 Merge pull request #7370 from oranagra/no_queue_in_aborted_multi 
Don't queue commands in an already aborted MULTI state
 2020-06-08 11:08:08 +02:00
Salvatore Sanfilippo
783735c915 Merge pull request #7369 from oranagra/no_reject_watch 
Avoid rejecting WATCH / UNWATCH, like MULTI/EXEC/DISCARD
 2020-06-08 11:04:26 +02:00
Salvatore Sanfilippo
0c539f6497 Merge pull request #7357 from soloestoy/bugfix-aof-keepttl 
AOF: append origin SET if no expire option
 2020-06-08 11:02:00 +02:00
Salvatore Sanfilippo
6fed3d8708 Merge pull request #7371 from oranagra/fix_disconnectSlaves 
fix disconnectSlaves, to try to free each slave.
 2020-06-08 10:43:51 +02:00
Salvatore Sanfilippo
4ea93c04ff Merge pull request #7352 from soloestoy/donot-free-protected-client-when-blocking 
donot free protected client in freeClientsInAsyncFreeQueue()
 2020-06-08 10:43:17 +02:00
Oran Agra
25ae9c15ce fix disconnectSlaves, to try to free each slave. 
the recent change in that loop (iteration rather than waiting for it to
be empty) was intended to avoid an endless loop in case some slave would
refuse to be freed.

but the lookup of the first client remained, which would have caused it
to try the first one again and again instead of moving on.
 2020-06-08 09:50:06 +03:00
Oran Agra
0bf6face8a Don't queue commands in an already aborted MULTI state 2020-06-08 09:43:10 +03:00
Oran Agra
b64641f995 Avoid rejecting WATCH / UNWATCH, like MULTI/EXEC/DISCARD 
Much like MULTI/EXEC/DISCARD, the WATCH and UNWATCH are not actually
operating on the database or server state, but instead operate on the
client state. the client may send them all in one long pipeline and check
all the responses only at the end, so failing them may lead to a
mismatch between the client state on the server and the one on the
client end, and execute the wrong commands (ones that were meant to be
discarded)

the watched keys are not actually stored in the client struct, but they
are in fact part of the client state. for instance, they're not cleared
or moved in SWAPDB or FLUSHDB.
 2020-06-08 09:16:32 +03:00
John Sully
ad4d76422c Some valgrind fixes 
Former-commit-id: ba974c9810dc30b6b5bd21c4bcc6cc5b219f7ea4
 2020-06-07 23:20:10 -04:00
John Sully
f252cc9e41 fix valgrind false error 
Former-commit-id: 9817db7eddf06becd151e5fef7cc6808777c49cb
 2020-06-07 20:17:47 -04:00
xhe
4c5eb1cac1 return the correct proto version 
HELLO should return the current proto version, while the code hardcoded
3
 2020-06-07 13:34:55 +08:00
Kevin Fwu
ce014c2a3b Fix TLS certificate loading for chained certificates. 
This impacts client verification for chained certificates (such as Lets
Encrypt certificates). Client Verify requires the full chain in order to
properly verify the certificate.
 2020-06-06 11:44:21 +02:00
antirez
5fe90c04a1 Merge branch 'unstable' of github.com:/antirez/redis into unstable 2020-06-06 11:44:05 +02:00
antirez
bec1ac9899 Revert "Implements sendfile for redis." 
This reverts commit 5675053269b0cbc2cf525c99321c96b7c2b39abe.
 2020-06-06 11:43:29 +02:00
antirez
8bfdba1e08 Revert "avoid using sendfile if tls-replication is enabled" 
This reverts commit 13bbd165e87923558952203d310e9ad053d4d7c0.
 2020-06-06 11:43:29 +02:00
Liu Zhen
123d94ffc0 fix clusters mixing accidentally by gossip 
`clusterStartHandshake` will start hand handshake
and eventually send CLUSTER MEET message, which is strictly prohibited
in the REDIS CLUSTER SPEC.
Only system administrator can initiate CLUSTER MEET message.
Futher, according to the SPEC, rather than IP/PORT pairs, only nodeid
can be trusted.
 2020-06-06 11:43:29 +02:00
antirez
a5f0fa7f7a Fix handling of special chars in ACL LOAD. 
Now it is also possible for ACL SETUSER to accept empty strings
as valid operations (doing nothing), so for instance

    ACL SETUSER myuser ""

Will have just the effect of creating a user in the default state.

This should fix #7329.
 2020-06-06 11:43:29 +02:00
antirez
8b15ba933d Revert "Implements sendfile for redis." 
This reverts commit 5675053269b0cbc2cf525c99321c96b7c2b39abe.
 2020-06-06 11:42:45 +02:00
antirez
3b452cac99 Revert "avoid using sendfile if tls-replication is enabled" 
This reverts commit 13bbd165e87923558952203d310e9ad053d4d7c0.
 2020-06-06 11:42:41 +02:00
John Sully
779023beff Ensure CRON jobs run in a clean environment 
Former-commit-id: c6dce838b7cc94e115fd73a64dda663f0a2c28c5
 2020-06-05 21:35:47 -04:00
zhaozhao.zz
cfc3e6f55d AOF: append origin SET if no expire option 2020-06-03 17:55:18 +08:00
zhaozhao.zz
14dc507de2 donot free protected client in freeClientsInAsyncFreeQueue 
related #7234
 2020-06-02 11:48:14 +08:00
John Sully
64adf015b1 Undecorated new is OK to use 
Former-commit-id: 5b885bb1649805f6a2edb8d28edd1447bb6c4843
 2020-06-01 23:36:01 -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
Salvatore Sanfilippo
a179552f8d Merge pull request #7334 from kevin-fwu/tls_load_chained_certs 
Fix TLS certificate loading for chained certificates.
 2020-05-31 14:32:56 +02:00
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
antirez
6df70da5ba Merge branch 'unstable' of github.com:/antirez/redis into unstable 2020-05-29 11:09:15 +02:00
antirez
7bb3ef2ec6 Fix handling of special chars in ACL LOAD. 
Now it is also possible for ACL SETUSER to accept empty strings
as valid operations (doing nothing), so for instance

    ACL SETUSER myuser ""

Will have just the effect of creating a user in the default state.

This should fix #7329.
 2020-05-29 11:07:13 +02:00
Salvatore Sanfilippo
fa3e801afe Merge pull request #7330 from liuzhen/unstable 
fix clusters mixing accidentally by gossip
 2020-05-29 10:15:26 +02:00
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