20963 Commits

Author SHA1 Message Date
John Sully
1b121723e2 Don't prefetch when lock contention is low, it increases latency
Former-commit-id: 0d21614e0e5aba28acd364231823d51a3073081f
2021-04-12 03:41:53 +00:00
John Sully
3ed9963363 Reduce P99 latency with async rehash
Former-commit-id: 7ea4c26fd82c0cdfa377183083f38a43336e480b
2021-04-12 03:39:13 +00:00
John Sully
e6782d8f1b Reduce P99 latency with async rehash
Former-commit-id: 7ea4c26fd82c0cdfa377183083f38a43336e480b
2021-04-12 03:39:13 +00:00
John Sully
aa64a260fc Make prefetch configurable
Former-commit-id: 3b660347d70cc25d57119080bd43fb4671e36488
2021-04-12 03:38:57 +00:00
John Sully
df66c598e9 Make prefetch configurable
Former-commit-id: 3b660347d70cc25d57119080bd43fb4671e36488
2021-04-12 03:38:57 +00:00
John Sully
da545be5b5 Make prefetch more aggressive
Former-commit-id: 9cb0be197fb96d7fb3cb697c19b5be313b01337a
2021-04-12 03:24:05 +00:00
John Sully
e541d82e1f Make prefetch more aggressive
Former-commit-id: 9cb0be197fb96d7fb3cb697c19b5be313b01337a
2021-04-12 03:24:05 +00:00
John Sully
5ccaa9265c Don't prefetch when lock contention is low, it increases latency
Former-commit-id: 9b2629f6a20368cec8e55f0d006f3a67c8b770b7
2021-04-12 03:23:49 +00:00
John Sully
c070f6ece2 Don't prefetch when lock contention is low, it increases latency
Former-commit-id: 9b2629f6a20368cec8e55f0d006f3a67c8b770b7
2021-04-12 03:23:49 +00:00
John Sully
2d2ae90f30 Reduce P99 latency with async rehash
Former-commit-id: 6c045837c7cf92dc92be35465229b482e09e46d2
2021-04-12 03:22:22 +00:00
John Sully
5b53c8e88e Reduce P99 latency with async rehash
Former-commit-id: 6c045837c7cf92dc92be35465229b482e09e46d2
2021-04-12 03:22:22 +00:00
Wang Yuan
8c5634f4eb Fix wrong check for aof fsync and handle aof fsync errno (#8751)
The bio aof fsync fd may be closed by main thread (AOFRW done handler)
and even possibly reused for another socket, pipe, or file.
This can can an EBADF or EINVAL fsync error, which will lead to -MISCONF errors failing all writes.
We just ignore these errno because aof fsync did not really fail.

We handle errno when fsyncing aof in bio, so we could know the real reason
when users get -MISCONF Errors writing to the AOF file error

Issue created with #8419
2021-04-11 08:14:31 +03:00
Wang Yuan
a0e19e3cf1
Fix wrong check for aof fsync and handle aof fsync errno (#8751)
The bio aof fsync fd may be closed by main thread (AOFRW done handler)
and even possibly reused for another socket, pipe, or file.
This can can an EBADF or EINVAL fsync error, which will lead to -MISCONF errors failing all writes.
We just ignore these errno because aof fsync did not really fail.

We handle errno when fsyncing aof in bio, so we could know the real reason
when users get -MISCONF Errors writing to the AOF file error

Issue created with #8419
2021-04-11 08:14:31 +03:00
yjph
2fa3e1d999 Fix the display of make install (#8667) 2021-04-10 21:25:53 +03:00
yjph
cd03e293c3
Fix the display of make install (#8667) 2021-04-10 21:25:53 +03:00
John Sully
c92b90eb91 DO not set the timethread as high priority, it can starve the server threads
Former-commit-id: faeac65371af9d6b6effe0886bcbdefaec24ad6d
2021-04-09 01:06:24 +00:00
John Sully
d58a415e29 DO not set the timethread as high priority, it can starve the server threads
Former-commit-id: faeac65371af9d6b6effe0886bcbdefaec24ad6d
2021-04-09 01:06:24 +00:00
John Sully
e1adedb1fa DO not set the timethread as high priority, it can starve the server threads
Former-commit-id: 5355207a0ad3f1a191fa06450490c792caa76e87
2021-04-09 01:05:55 +00:00
John Sully
788119a964 DO not set the timethread as high priority, it can starve the server threads
Former-commit-id: 5355207a0ad3f1a191fa06450490c792caa76e87
2021-04-09 01:05:55 +00:00
John Sully
69f7a194bc Make prefetch configurable
Former-commit-id: 16996c048042bd3799c8051645bbe7c54137d54c
2021-04-08 19:52:38 +00:00
John Sully
9d0001ae7c Make prefetch configurable
Former-commit-id: 16996c048042bd3799c8051645bbe7c54137d54c
2021-04-08 19:52:38 +00:00
John Sully
571718f774 Avoid posting unnecessary async tasks
Former-commit-id: 08f63fdfd6c78388bba093ec2edda7d77fc9173e
2021-04-07 17:41:20 +00:00
John Sully
5d924bf16a Avoid posting unnecessary async tasks
Former-commit-id: 08f63fdfd6c78388bba093ec2edda7d77fc9173e
2021-04-07 17:41:20 +00:00
John Sully
635ddad7fc Fix issue where GC is not free'd until a BGSAVE
Former-commit-id: e19350005b571591876e49219bfca75e905604a5
2021-04-07 17:30:59 +00:00
John Sully
c0e80ffcda Fix issue where GC is not free'd until a BGSAVE
Former-commit-id: e19350005b571591876e49219bfca75e905604a5
2021-04-07 17:30:59 +00:00
John Sully
55a118ecd0 Reduce connection latency for cluster clients
Former-commit-id: f1acee99bdaee4faf9e18cdf4734d51a73e78a41
2021-04-07 15:49:58 +00:00
John Sully
86a5b2fbc7 Reduce connection latency for cluster clients
Former-commit-id: f1acee99bdaee4faf9e18cdf4734d51a73e78a41
2021-04-07 15:49:58 +00:00
John Sully
eb818d4c00 Fix build break from merge conflicts
Former-commit-id: 5052388bd43b0e42430fd741fb2b403544684246
2021-04-07 15:25:22 +00:00
John Sully
1bf5477602 Fix build break from merge conflicts
Former-commit-id: 5052388bd43b0e42430fd741fb2b403544684246
2021-04-07 15:25:22 +00:00
John Sully
ca13d87204 Merge branch 'keydbpro' of https://gitlab.eqalpha.com/external-collab/keydb-pro-6 into keydbpro_collab
Former-commit-id: e077756b140855cb5a7e60567abf845dbfbc162d
2021-04-07 15:14:59 +00:00
John Sully
d832ebb221 Merge branch 'keydbpro' of https://gitlab.eqalpha.com/external-collab/keydb-pro-6 into keydbpro_collab
Former-commit-id: e077756b140855cb5a7e60567abf845dbfbc162d
2021-04-07 15:14:59 +00:00
Wen Hui
dfbfe4c558 use getPositiveLongFromObjectOrReply for positive check of args (#8750)
just a cleanup
2021-04-07 10:28:53 +03:00
Wen Hui
d96f47cf06
use getPositiveLongFromObjectOrReply for positive check of args (#8750)
just a cleanup
2021-04-07 10:28:53 +03:00
Yang Bodong
79793c3d55 Fix out of range confusing error messages (XAUTOCLAIM, RPOP count) (#8746)
Fix out of range error messages to be clearer (avoid mentioning 9223372036854775807)
* Fix XAUTOCLAIM COUNT option confusing error msg
* Fix other RPOP and alike error message to mention positive
2021-04-07 10:01:28 +03:00
Yang Bodong
4c14e8668c
Fix out of range confusing error messages (XAUTOCLAIM, RPOP count) (#8746)
Fix out of range error messages to be clearer (avoid mentioning 9223372036854775807)
* Fix XAUTOCLAIM COUNT option confusing error msg
* Fix other RPOP and alike error message to mention positive
2021-04-07 10:01:28 +03:00
jsully
c168c8fbbf Merge branch 'flash_read_perf' into 'keydbpro'
Flash read perf

See merge request external-collab/keydb-pro-6!1

Former-commit-id: f6d7277f307b78f079dd99bf532cd26a7b62018d
2021-04-07 06:12:55 +00:00
jsully
59a9e93abb Merge branch 'flash_read_perf' into 'keydbpro'
Flash read perf

See merge request external-collab/keydb-pro-6!1

Former-commit-id: f6d7277f307b78f079dd99bf532cd26a7b62018d
2021-04-07 06:12:55 +00:00
Cloud User
6601905d8c Make high pri time thread configurable
Former-commit-id: 7e94207765d2166f46792aea0919786f3f30d7b3
2021-04-06 22:39:20 +00:00
Cloud User
6684f150d6 Make high pri time thread configurable
Former-commit-id: 7e94207765d2166f46792aea0919786f3f30d7b3
2021-04-06 22:39:20 +00:00
Oran Agra
7af6fff7b9 update help.h (#8747) 2021-04-06 12:42:18 +03:00
Oran Agra
4de1031d8a
update help.h (#8747) 2021-04-06 12:42:18 +03:00
yjph
1ea4f128d3 Update the location information in some URLs (#8595) 2021-04-06 12:29:02 +03:00
yjph
073598ed8f
Update the location information in some URLs (#8595) 2021-04-06 12:29:02 +03:00
Bonsai
15e2b68214 Fix: server will crash if rdbload or rdbsave method is not provided in module (#8670)
With this fix, module data type registration will fail if the load or save callbacks are not defined, or the optional aux load and save callbacks are not either both defined or both missing.
2021-04-06 12:09:36 +03:00
Bonsai
808f3004f0
Fix: server will crash if rdbload or rdbsave method is not provided in module (#8670)
With this fix, module data type registration will fail if the load or save callbacks are not defined, or the optional aux load and save callbacks are not either both defined or both missing.
2021-04-06 12:09:36 +03:00
Huang Zhw
f5cc88e65d Fix "default" and overwritten / reset users will not have pubsub channels permissions by default. (#8723)
Background:
Redis 6.2 added ACL control for pubsub channels (#7993), which were supposed
to be permissive by default to retain compatibility with redis 6.0 ACL. 
But due to a bug, only newly created users got this `acl-pubsub-default` applied,
while overwritten (updated) users got reset to `resetchannels` (denied).

Since the "default" user exists before loading the config file,
any ACL change to it, results in an update / overwrite.

So when a "default" user is loaded from config file or include ACL
file with no channels related rules, the user will not have any
permissions to any channels. But other users will have default
permissions to any channels.

When upgraded from 6.0 with config rewrite, this will lead to
"default" user channels permissions lost.
When users are loaded from include file, then call "acl load", users
will also lost channels permissions.

Similarly, the `reset` ACL rule, would have reset the user to be denied
access to any channels, ignoring `acl-pubsub-default` and breaking
compatibility with redis 6.0.

The implication of this fix is that it regains compatibility with redis 6.0,
but breaks compatibility with redis 6.2.0 and 2.0.1. e.g. after the upgrade,
the default user will regain access to pubsub channels.

Other changes:
Additionally this commit rename server.acl_pubusub_default to
server.acl_pubsub_default and fix typo in acl tests.
2021-04-05 23:13:20 +03:00
Huang Zhw
3b74b55084
Fix "default" and overwritten / reset users will not have pubsub channels permissions by default. (#8723)
Background:
Redis 6.2 added ACL control for pubsub channels (#7993), which were supposed
to be permissive by default to retain compatibility with redis 6.0 ACL. 
But due to a bug, only newly created users got this `acl-pubsub-default` applied,
while overwritten (updated) users got reset to `resetchannels` (denied).

Since the "default" user exists before loading the config file,
any ACL change to it, results in an update / overwrite.

So when a "default" user is loaded from config file or include ACL
file with no channels related rules, the user will not have any
permissions to any channels. But other users will have default
permissions to any channels.

When upgraded from 6.0 with config rewrite, this will lead to
"default" user channels permissions lost.
When users are loaded from include file, then call "acl load", users
will also lost channels permissions.

Similarly, the `reset` ACL rule, would have reset the user to be denied
access to any channels, ignoring `acl-pubsub-default` and breaking
compatibility with redis 6.0.

The implication of this fix is that it regains compatibility with redis 6.0,
but breaks compatibility with redis 6.2.0 and 2.0.1. e.g. after the upgrade,
the default user will regain access to pubsub channels.

Other changes:
Additionally this commit rename server.acl_pubusub_default to
server.acl_pubsub_default and fix typo in acl tests.
2021-04-05 23:13:20 +03:00
Huang Zhw
84414bb33f redis-cli --bigkeys / memkeys, report detailed error on dbsize failure (#8740)
When DBSIZE failed (e.g. on AUTH error), the printed error didn't reflect the reason.
2021-04-05 08:30:41 +03:00
Huang Zhw
a3da3e592d
redis-cli --bigkeys / memkeys, report detailed error on dbsize failure (#8740)
When DBSIZE failed (e.g. on AUTH error), the printed error didn't reflect the reason.
2021-04-05 08:30:41 +03:00
Sokolov Yura
0d7a17b252 Add cluster-allow-replica-migration option. (#5285)
Previously (and by default after commit) when master loose its last slot
(due to migration, for example), its replicas will migrate to new last slot
holder.

There are cases where this is not desired:
* Consolidation that results with removed nodes (including the replica, eventually).
* Manually configured cluster topologies, which the admin wishes to preserve.

Needlessly migrating a replica triggers a full synchronization and can have a negative impact, so
we prefer to be able to avoid it where possible.

This commit adds 'cluster-allow-replica-migration' configuration option that is
enabled by default to preserve existed behavior. When disabled, replicas will
not be auto-migrated.

Fixes #4896

Co-authored-by: Oran Agra <oran@redislabs.com>
2021-04-04 09:43:24 +03:00