Changes in cluster.h

08db7092 - Cluster: clarify node->slave may be NULL.

Thu, 14 Jan 2016 10:58:31 +0000

Cluster: clarify node->slave may be NULL. List of files: /redis-3.2.3/src/cluster.h

6b8a6c2a - Cluster: replica migration with delay.

Fri, 11 Dec 2015 08:19:06 +0000

Cluster: replica migration with delay.We wait a fixed amount of time (5 seconds currently) much greater thanthe usual Cluster node to node communication latency, before migrating.This way when a failover occurs, before detecting the new master as atarget for migration, we give the time to its natural slaves (the slavesof the failed over master) to announce they switched to the new master,preventing an useless migration operation. List of files: /redis-3.2.3/src/cluster.h

db65f6d3 - Fix replicas migration by adding a new flag.

Wed, 09 Dec 2015 21:52:53 +0000

Fix replicas migration by adding a new flag.Some time ago I broken replicas migration (reported in #2924).The idea was to prevent masters without replicas from getting replicasbecause of replica migration, I remember it to create issues with tests,but there is no clue in the commit message about why it was soundesirable.However my patch as a side effect totally ruined the concept of replicasmigration since we want it to work also for instances that, technically,never had slaves in the past: promoted slaves.So now instead the ability to be targeted by replicas migration, is anew flag "migrate-to". It only applies to masters, and is set in thefollowing two cases:1. When a master gets a slave, it is set.2. When a slave turns into a master because of fail over, it is set.This way replicas migration targets are only masters that used to haveslaves, and slaves of masters (that used to have slaves... obviously)and are promoted.The new flag is only internal, and is never exposed in the output norpersisted in the nodes configuration, since all the information tohandle it are implicit in the cluster configuration we already have. List of files: /redis-3.2.3/src/cluster.h

3325a9b1 - RDMF: more names updated.

Mon, 27 Jul 2015 12:55:45 +0000

RDMF: more names updated. List of files: /redis-3.2.3/src/cluster.h

32f80e2f - RDMF: More consistent define names.

Mon, 27 Jul 2015 07:41:48 +0000

RDMF: More consistent define names. List of files: /redis-3.2.3/src/cluster.h

554bd0e7 - RDMF: use client instead of redisClient, like Disque.

Sun, 26 Jul 2015 13:20:46 +0000

RDMF: use client instead of redisClient, like Disque. List of files: /redis-3.2.3/src/cluster.h

9b7f8b1c - Cluster: redirection refactoring + handling of blocked clients.

Tue, 24 Mar 2015 10:07:10 +0000

Cluster: redirection refactoring + handling of blocked clients.There was a bug in Redis Cluster caused by clients blocked in a blockinglist pop operation, for keys no longer handled by the instance, orin a condition where the cluster became down after the client blocked.A typical situation is:1) BLPOP 02) hash slot is resharded to another master.The client will block forever int this case.A symmentrical non-cluster-specific bug happens when an instance isturned from master to slave. In that case it is more serious since thiswill desynchronize data between slaves and masters. This other bug wasdiscovered as a side effect of thinking about the bug explained andfixed in this commit, but will be fixed in a separated commit. List of files: /redis-3.2.3/src/cluster.h

25c0f5ac - Cluster: better cluster state transiction handling.

Fri, 20 Mar 2015 08:57:10 +0000

Cluster: better cluster state transiction handling.Before we relied on the global cluster state to make sure all the hashslots are linked to some node, when getNodeByQuery() is called. Sofinding the hash slot unbound was checked with an assertion. Howeverthis is fragile. The cluster state is often updated in theclusterBeforeSleep() function, and not ASAP on state change, so it mayhappen to process clients with a cluster state that is 'ok' but yetcertain hash slots set to NULL.With this commit the condition is also checked in getNodeByQuery() andreported with a identical error code of -CLUSTERDOWN but slightlydifferent error message so that we have more debugging clue in thefuture.Root cause of issue #2288. List of files: /redis-3.2.3/src/cluster.h

5b0f4a83 - Cluster: clusterMsgDataGossip structure, explict padding + minor stuff.

Tue, 13 Jan 2015 09:20:49 +0000

Cluster: clusterMsgDataGossip structure, explict padding + minor stuff.Also explicitly set version to 0, add a protocol version define, improvecomments in the gossip structure.Note that the structure layout is the same after the change, we are justmaking the padding explicit with an additional not used 16 bits field.So this commit is still able to talk with the previous versions ofcluster nodes. List of files: /redis-3.2.3/src/cluster.h

669aa2a2 - Cluster PUBLISH message: fix totlen count.

Fri, 28 Nov 2014 09:03:07 +0000

Cluster PUBLISH message: fix totlen count.bulk_data field size was not removed from the count. It is not possibleto declare it simply as 'char bulk_data[]' since the structure is nestedinto another structure. List of files: /redis-3.2.3/src/cluster.h

edb3987a - Cluster: more chatty slaves when failover is stalled.

Tue, 07 Oct 2014 07:51:55 +0000

Cluster: more chatty slaves when failover is stalled. List of files: /redis-3.2.3/src/cluster.h

12d0195b - Clean up text throughout project

Thu, 31 Jul 2014 18:51:05 +0000

Clean up text throughout project - Remove trailing newlines from redis.conf - Fix comment misspelling - Clarifies zipEncodeLength usage and a C API mention (#1243, #1242) - Fix cluster typos (inspired by @papanikge #1507) - Fix rewite -> rewrite in a few places (inspired by #682)Closes #1243, #1242, #1507 List of files: /redis-3.2.3/src/cluster.h

c89afc8e - Cluster: new option to work with partial slots coverage.

Wed, 17 Sep 2014 09:10:09 +0000

Cluster: new option to work with partial slots coverage. List of files: /redis-3.2.3/src/cluster.h

119a7000 - Use modern typedef form in cluster.h.

Mon, 25 Aug 2014 08:42:18 +0000

Use modern typedef form in cluster.h. List of files: /redis-3.2.3/src/cluster.h

39603a7e - Cluster: slave validity factor is now user configurable.

Thu, 22 May 2014 14:57:47 +0000

Cluster: slave validity factor is now user configurable.Check the commit changes in the example redis.conf for more information. List of files: /redis-3.2.3/src/cluster.h

8f52173b - Cluster: last_vote_epoch -> lastVoteEpoch.

Thu, 27 Mar 2014 14:01:24 +0000

Cluster: last_vote_epoch -> lastVoteEpoch.Use cammel case for epochs that are persisted on disk. List of files: /redis-3.2.3/src/cluster.h

0f1f2578 - Cluster: better timeout and retry time for failover.

Mon, 10 Mar 2014 08:57:52 +0000

Cluster: better timeout and retry time for failover.When node-timeout is too small, in the order of a few milliseconds,there is no way the voting process can terminate during that time, so weset a lower limit for the failover timeout of two seconds.The retry time is set to two times the failover timeout time, so it isat least 4 seconds. List of files: /redis-3.2.3/src/cluster.h

36676c23 - Redis Cluster: support for multi-key operations.

Fri, 07 Mar 2014 12:19:09 +0000

Redis Cluster: support for multi-key operations. List of files: /redis-3.2.3/src/cluster.h

385c25f7 - Remove redundant IP length definition

Wed, 26 Feb 2014 23:05:12 +0000

Remove redundant IP length definitionREDIS_CLUSTER_IPLEN had the same value asREDIS_IP_STR_LEN. They were both #define'dto the same INET6_ADDRSTRLEN. List of files: /redis-3.2.3/src/cluster.h

d769cad4 - Fix IP representation in clusterMsgDataGossip

Tue, 25 Feb 2014 21:02:28 +0000

Fix IP representation in clusterMsgDataGossip List of files: /redis-3.2.3/src/cluster.h