raboof/pekko
1
0
Fork 0
Code Issues Pull requests Projects Releases 6 Packages Wiki Activity Actions 11
18293 commits 1 branch 31 tags 268 MiB
Commit graph

62 commits

Author SHA1 Message Date
Patrik Nordwall
5ebdd79bee =cls increase the delay in the graceful shutdown example 2015-12-21 09:54:14 +01:00
Roland Kuhn
6a2d4e2ce6 Merge pull request #19226 from akka/wip-clarify-sharding-role-patriknw 
=cls Clarify the sharding role setting
 2015-12-20 18:49:36 +01:00
Patrik Nordwall
f5ed085179 =cls improve the graceful shutdown example 2015-12-18 11:39:52 +01:00
Patrik Nordwall
d44120b726 =cls Clarify the sharding role setting 2015-12-18 11:15:06 +01:00
Patrik Nordwall
a6fd7b448f =cls #18978 Lazy startup of shards when rememberEntities=false 
* and don't populate the unallocatedShards Set in the State
  when rememberEntities=false
 2015-11-27 10:09:44 +01:00
Patrik Nordwall
d952302edc Merge pull request #18984 from msosnicki/msosnicki-18945 
Issue #18945. Fix HandOff problem where Shards may not be killed.
 2015-11-25 09:22:57 +01:00
Marcin Sosnicki
e77bb2fa45 Issue #18945. Fix HandOff problem where Shards may not be killed. 
Issue #18945. Fix HandOff problem where Shards may not be killed.
 2015-11-23 20:36:25 +00:00
Patrik Nordwall
27995af79f =cls #18722 fix DDataShardCoordinator init 
* the become logic was wrong when watchStateActors triggers an immediate
  state update
 2015-11-18 16:13:58 +01:00
Krzysztof Bochenek
5c418efef2 =cls #18762 fix graceful shutdown of empty region 2015-11-11 11:58:43 +01:00
drewhk
11bf8a67d0 Merge pull request #18543 from johanandren/wip-17695-api-for-sharding-state-retrieval-johanandren 
+clu #17695 add a way to inspect the current sharding state
 2015-11-11 11:19:50 +01:00
Patrik Nordwall
1e36e5e187 Merge pull request #18746 from akka/wip-18554-singleton-startup-patriknw 
=clu #18554 Make oldest assignment deterministic when joining
 2015-11-06 14:48:57 +01:00
Patrik Nordwall
c7c187f6b7 =clu replace Set -- with diff and ++ with union 
* better performance according to
  https://docs.google.com/presentation/d/1Qjryxoe-fYEM8ZPhM-98LKfbhnRcn5eAEMNlVVnixsA/pub
 2015-11-06 14:48:17 +01:00
Johan Andrén
4abbc8db50 +clu #17695 add a way to inspect the current sharding state 
Two new message pairs:
`GetShardRegionState`/`CurrentShardRegionState` allows for querying a region for it's current shards and the current `EntityIds` of it
`GetClusterShardingStats`/`ClusterShardingStats` allows for querying the entire cluster for a summary of
the number of entitites alive in each region and shard.
 2015-11-02 08:56:09 +01:00
Martynas Mickevičius
4ffe267835 =pro verify BC against all binary compatible versions 2015-10-23 09:10:25 +03:00
Patrik Nordwall
9380983d3c =clu #18554 Make oldest assignment deterministic when joining 
* the reported issue is fixed by the immediate leaderActions
  (moving to Up)  when joining the first node to itself
* the other changes are precautions just in case
 2015-10-21 07:53:14 +02:00
Patrik Nordwall
48754bb33f =pro #18447 Check binary compatibility with 2.4.0 2015-09-30 11:31:11 +02:00
Patrik Nordwall
0cdd9af3c4 +per #18485 Add cluster sharding data removal utility 2015-09-17 14:07:08 +02:00
Konrad Malawski
c57b4e24c8 Merge pull request #18445 from akka/wip-18370-sharding-supervision-patriknw 
=cls #18370 Document supervision for Cluster Sharding
 2015-09-16 12:59:37 +02:00
Patrik Nordwall
4e2b8190a3 =cls #18370 Document supervision for Cluster Sharding 2015-09-10 15:35:26 +02:00
Patrik Nordwall
e5159eb764 =cls #18176 Harden ClusterShardingLeavingSpec 
In logs it is clear that the fourth node is moved to Up,
but it takes more than 5 sec to disseminate that info
 2015-09-09 14:36:08 +02:00
Patrik Nordwall
c9662d8083 Merge pull request #18324 from akka/wip-15646-sharding-initial-watch-patriknw 
=cls #15646 Optimize the initial watch in shard coordinator
 2015-09-04 12:02:17 +02:00
Patrik Nordwall
4b6382f780 Merge pull request #18363 from akka/wip-13783-embedded-protobuf-patriknw 
=all #13783 replace protobuf dependency by embedded version
 2015-09-04 11:30:01 +02:00
Patrik Nordwall
bfde1eff19 =clu #18337 Disable down-removal-margin by default 
For manual downing it is not needed. For auto-down it doesn't add any extra safety, since that
is not handling network partitions anyway.

The setting is still useful if you implement downing strategies that handle network partitions,
e.g. by keeping the larger side of the partition and shutting down the smaller side.
 2015-09-04 11:28:33 +02:00
Roland Kuhn
c3ecb87a65 =all #13783 replace protobuf dependency by embedded version 
- created new subproject akka-protobuf (and added COPYING and LICENSE)
- renamed com.google.protobuf -> akka.protobuf everywhere
- also added such renaming step to the results of protoc compilation in
  project/Protobuf.scala
- had to include transcriptions of Netty’s ProtobufEncoder/Decoder to
  make multi-node-testkit compile again
 2015-09-04 09:06:50 +02:00
Brendan McAdams
607ea3e59a Fixed a spelling error in a cluster sharding error message when coordinator is not contactable. 2015-08-27 14:30:57 -07:00
Patrik Nordwall
bc48872873 =cls #15646 Optimize the initial watch in shard coordinator 
Two improvements to the coordinator startup (state recovery) that
should make it operational faster and reduce the amount of lost messages
during startup.

* Let the quick (those not involving failure detection) Terminated messages
  be processed before starting to reply to GetShardHome.
* Consider regions that don't belong to the current cluster
  to be terminated.
 2015-08-27 18:45:32 +02:00
Patrik Nordwall
ba135e8e34 =clt #18314 Reduce message lost when region terminates 
The new akka.cluster.down-removal-margin comes into play.
During that period messages are still routed to the old location, even though we have got the Terminated message.

We can reduce (best effort) the message loss by not replying to GetShardHome during the period.
 2015-08-27 10:20:10 +02:00
Ostapenko Evgeniy
9926658f7e =cls #18287 failed: ClusterShardingSpec ddata mode 2015-08-25 19:11:25 +03:00
Patrik Nordwall
50d7b1553a =cls #17846 Use provided scope for the distributed-data dependency 2015-08-21 14:36:52 +02:00
Patrik Nordwall
b9fecfd53b Merge pull request #18257 from akka/wip-17699-DeadLetterSuppression-patriknw 
=cls #17699 Add some DeadLetterSuppression
 2015-08-21 08:46:30 +02:00
Patrik Nordwall
834a84fdd6 =cls #17846 Add experimental note to ddata mode docs 2015-08-20 19:07:41 +02:00
Ostapenko Evgeniy
6814d08ef1 =cls #17846 Use CRDTs instead of PersistentActor to remember the state of the ShardCoordinator #17871 2015-08-20 13:36:37 +03:00
Patrik Nordwall
775a2ca40c =cls #17699 Add some DeadLetterSuppression 2015-08-19 15:38:05 +02:00
Konrad Malawski
4e59f0ea35 +per #18137 EventAdapter => can be Read / Write or both 2015-08-19 15:26:30 +02:00
Patrik Nordwall
46811ca1ce =cls #17908 Protobuf serializer for cluster sharding 2015-08-17 16:11:57 +02:00
Patrik Nordwall
c9f0021297 =cls #16090 Add more logging for buffering and unavailable coordinator 2015-08-13 10:48:43 +02:00
Konrad Malawski
86c00d4716 !per +act #17842 move BackoffSupervisor to akka.pattern 2015-07-08 16:45:23 +02:00
Patrik Nordwall
89f17ddfd0 =cls #17447 Split Cluster Sharding docs into java/scala 2015-06-30 16:39:31 +02:00
Patrik Nordwall
2832dd55c5 !clt, cls #17866 Use systemActorOf for exension actors 
* ClusterSharding
* ClusterClientReceptionist
* dispatcher config, since deployment config can't be used
  for system actors
 2015-06-30 16:37:34 +02:00
Patrik Nordwall
985510d6ac =cls #17261 Use persistent shard only when rememberEntities=true 2015-06-22 08:54:42 +02:00
Patrik Nordwall
d02b003628 =cls Split sharding classes into separate files 2015-06-22 08:09:10 +02:00
Roland Kuhn
0de9f0ff40 Merge pull request #17641 from kukido/kukido-spellings-normalization 
=doc #17329 Fixed and normalized spellings in ScalaDoc and comments
 2015-06-19 12:06:53 +02:00
Patrik Nordwall
2a88f4fb29 =clu Improve cluster downing 
* avoid using Down and Exiting member from being used for joining
* delay shut down of Down member until the information is spread
  to all reachable members, e.g. downing several nodes via one node
* akka.cluster.down-removal-margin setting
  Margin until shards or singletons that belonged to a
  downed/removed partition are created in surviving partition.
  Used by singleton and sharding.
* remove the retry count parameters/settings for singleton in
  favor of deriving those from the removal-margin
 2015-06-18 12:55:54 +02:00
Patrik Nordwall
6d26b3e591 !per Make persistent failures fatal 
* remove PersistentFailure and RecoveryFailure messages
* use stop instead of ActorKilledException
* adjust PersistentView
* adjust AtLeastOnceDeliveryFailureSpec
* adjust sharding
* add BackoffSupervisor
 2015-06-17 15:49:47 +02:00
Patrik Nordwall
5fab2b4521 !cls #16422 Rename shardResolver and idExtractor 2015-06-16 13:38:57 +02:00
Patrik Nordwall
70024298ac !cls #16422 Rename Entry to Entity in sharding 2015-06-11 10:00:43 +02:00
Patrik Nordwall
e8a5ad7292 !cls java8 Optional 2015-06-11 10:00:43 +02:00
Patrik Nordwall
25ba89a98b =cls #15614 Change persistenceId for sharding coordinator 2015-06-11 10:00:43 +02:00
Patrik Nordwall
8276420a89 =cls #15619 Use event counting instead of time based snapshot in sharding 2015-06-11 10:00:42 +02:00
Patrik Nordwall
c9a2447867 +cls #15330 Add GetCurrentRegions, for testing 2015-06-11 10:00:42 +02:00