Roland
80825d2433
fix TcpConnection’s PendingWrite.wantsAck
2013-02-15 14:29:37 +01:00
Endre Sándor Varga
acb927e870
Added missing return types in UdpFFMessage
2013-02-15 13:53:02 +01:00
Björn Antonsson
cd2b499b79
Java API for UdpConn
2013-02-15 13:11:35 +01:00
Roland
6d61a59a0f
add INTERNAL API markers and explicit return types
2013-02-15 13:11:35 +01:00
Endre Sándor Varga
933c93c05b
Added Java API for UdpFF
2013-02-15 13:11:35 +01:00
Roland
d71a541596
IO layer: add Java API and docs for TCP
2013-02-15 13:11:34 +01:00
Roland
fe5769695f
brown paper bag fix to the build
2013-02-14 16:04:36 +01:00
Roland
746adeaed7
add allTests pseudo-project for BC verification of Scala
2013-02-14 15:49:21 +01:00
Patrik Nordwall
9b5356f58c
Correct link in io docs
2013-02-14 08:29:54 +01:00
Rich Dougherty
a3b0e9375c
Merge pull request #1127 from akka/wip-3035-io-exceptions-rich
...
IOManager connect and accept failure handling. Fixes #3035
2013-02-13 15:58:50 -08:00
Patrik Nordwall
24bff5e5ca
Merge pull request #1131 from akka/wip-3041-ClusterDomainEventPublisherSpec-patriknw
...
Fix ClusterDomainEventPublisherSpec, see #3041
2013-02-13 13:11:26 -08:00
Patrik Nordwall
aad95c7e69
Fix ClusterDomainEventPublisherSpec, see #3041
...
* MemberRemoved is now published in postStop of ClusterDomainEventPublisher
* No need to shutdown the publisher in the test
2013-02-13 20:39:07 +01:00
Roland Kuhn
12de575139
Merge pull request #1129 from akka/wip-IO
...
Wip io
2013-02-13 08:29:51 -08:00
Patrik Nordwall
0ff5f2d19a
Merge pull request #1128 from akka/wip-3034-cluster-lars-patriknw
...
Create cluster scheduler from configured scheduler class, see #3034
2013-02-13 08:00:17 -08:00
Patrik Nordwall
6da6223e15
Merge pull request #1124 from akka/wip-3022-ssl-settings-patriknw
...
Restore ssl settings in reference.conf, see #3022
2013-02-13 07:59:15 -08:00
Roland Kuhn
ff1a92d939
Merge pull request #1126 from akka/wip-3029-receive-n-rich
...
Add receiveN to JavaTestKit. Fixes #3029
2013-02-13 07:48:14 -08:00
Patrik Nordwall
33bc2ed5f6
Merge pull request #1122 from akka/wip-3030-cluster-fault-handling-patriknw
...
Make cluster fault handling more robust, see #3030
2013-02-13 07:39:22 -08:00
Roland Kuhn
f334b1cfe5
Merge pull request #1098 from drewhk/wip-IO-UDP-fire-and-forget-drewhk
...
UDP implementation and generalized Selector
2013-02-13 07:25:25 -08:00
Roland Kuhn
13b5f5724f
Merge pull request #1125 from akka/wip-3026-camel-tests-∂π
...
ensure unique actor names in camel ProducerFeatureTest, see #3026
2013-02-13 07:09:33 -08:00
Patrik Nordwall
f706156fd7
Create cluster scheduler from configured scheduler class, see #3034
2013-02-13 15:37:27 +01:00
Rich Dougherty
acadb9e7cb
IOManager connect and accept failure handling. Fixes #3035
...
* Cherry-picked from release-2.1.
* Small changes: channel.socket -> sock
2013-02-13 21:16:54 +13:00
Rich Dougherty
b6c1527cbd
Add receiveN to JavaTestKit. Fixes #3029
2013-02-13 15:30:15 +13:00
Rich Dougherty
44808c17ed
Merge pull request #1095 from akka/wip-2874-router-doc-rich
...
Document router handling of PoisonPill messages. Fixes #2874
2013-02-12 16:31:04 -08:00
Patrik Nordwall
b349ad8d87
Nodes not part of cluster have marked the Gossip as seen, see #3031
...
* Problem may occur when joining member with same hostname:port again,
after downing.
* Reproduced with StressSpec exerciseJoinRemove with fixed port that
joins and shutdown several times.
* Real solution for this will be covered by ticket #2788 by adding
uid to member identifier, but as first step we need to support
this scenario with current design.
* Use unique node identifier for vector clock to avoid mixup of
old and new member instance.
* Support transition from Down to Joining in Gossip merge
* Don't gossip to unknown or unreachable members.
2013-02-12 21:55:08 +01:00
Patrik Nordwall
cab78e5174
Make cluster fault handling more robust, see #3030
...
* ClusterCoreDaemon and ClusterDomainEventPublisher can't be restarted
because the state would be obsolete.
* Add extra supervisor level for ClusterCoreDaemon and
ClusterDomainEventPublisher, which will shutdown the member
on failure in children.
* Publish the final removed state on postStop in
ClusterDomainEventPublisher. This also simplifies the removing
process.
2013-02-12 21:55:08 +01:00
Endre Sándor Varga
32d1a0072b
SimpleSender now works with the companion object
2013-02-12 13:08:41 +01:00
Rich Dougherty
3b98ba249b
Improved router documentation. Fixes #2874
...
* Edit of basic usage and design.
* Rearranged order of sections.
* Added more information on special message types.
* Added receiveN to JavaTestKit.
2013-02-12 12:25:41 +13:00
Viktor Klang (√)
b002bda23f
Merge pull request #1116 from akka/wip-3023-adding-docs-for-gradle-√
...
#3023 - Adding Gradle section to Getting Started docs and clarifying wh...
2013-02-11 15:20:20 -08:00
Björn Antonsson
94c0848fe3
Merge pull request #1123 from akka/wip-3027-LightArrayRevolverSchedulerSpec-failed-ban
...
Scheduler shutdown timeout needs to be longer on slow machines. See #3027
2013-02-11 11:25:59 -08:00
Patrik Nordwall
452275302b
Merge pull request #1120 from akka/wip-3028-dataflow-typo-rich
...
Typo in dataflow docs. Fixes #3028
2013-02-11 07:04:44 -08:00
Roland
5f25169095
ensure unique actor names in camel ProducerFeatureTest, see #3026
2013-02-11 15:25:25 +01:00
Patrik Nordwall
9ed701a75c
Restore ssl settings in reference.conf, see #3022
...
* Cleanup and addition of configuration tests for remote settings
2013-02-11 14:45:00 +01:00
Roland Kuhn
9915e3e263
Merge pull request #1118 from akka/wip-2981-remote-tests-∂π
...
move multi-node-testkit into its own module, see #2981
2013-02-11 05:38:33 -08:00
Roland
c3b363c9b3
move multi-node-testkit into its own module, see #2981
2013-02-11 14:37:37 +01:00
Roland Kuhn
29b274b9c0
Merge pull request #1119 from akka/wip-3016-dataflow-deps-∂π
...
remove actor-dep from dataflow, see #3016
2013-02-11 05:35:09 -08:00
Björn Antonsson
6621cda536
Scheduler shutdown timeout needs to be longer on slow machines. See #3027
2013-02-11 13:06:33 +01:00
Björn Antonsson
814ba1c02a
Merge pull request #1115 from akka/wip-2977-remotingspec-ssl-failure-ban
...
We can't set the SHA1PRNG random source reliably from within Akka. See #2977
2013-02-11 00:25:26 -08:00
Rich Dougherty
4fb48e29b4
Typo in dataflow docs. Fixes #3028
2013-02-11 12:23:14 +13:00
Roland
9949913fc4
remove actor-dep from dataflow, see #3016
2013-02-10 16:25:01 -05:00
Endre Sándor Varga
63264e847a
Second iteration of documentation
...
- Added basic UDP and TCP sections
2013-02-10 18:34:29 +01:00
Patrik Nordwall
1fcd035cf8
Merge pull request #1114 from akka/wip-2827-remote-api-patriknw
...
Polish API/SPI of remoting, see #2827
2013-02-10 08:49:11 -08:00
Patrik Nordwall
323e5c80b5
Polish the API/SPI of remoting, see #2827
...
* Changed TransportAdapterProvider to support java impl
* Verified java impl of AbstractTransportAdapter and
ActorTransportAdapter
* Privatized things that should not be public api
* Consistent usage of INTERNAL API marker in scaladoc
* Added some missing doc in conf
* Added missing SerialVersionUID
2013-02-10 17:47:43 +01:00
Patrik Nordwall
af7ca554c9
Merge pull request #1112 from akka/wip-3017-singleton-bug-patriknw
...
Hardening of another corner case in cluster singleton, see #3017
2013-02-10 08:07:32 -08:00
Endre Sándor Varga
bddcf9ba8c
First iteration of documentation
2013-02-10 16:46:59 +01:00
Endre Sándor Varga
e9da097621
Fixed according to revew comments
2013-02-10 13:52:52 +01:00
drewhk
60d22802b4
Merge pull request #1069 from drewhk/wip-new-remoting-docs-drewhk
...
Updated documentation to reflect changes in remoting
2013-02-10 04:02:28 -08:00
Endre Sándor Varga
c4abbd95bd
Updated documentation to reflect changes in remoting
2013-02-10 13:00:30 +01:00
Viktor Klang
07f3ee3d19
#3023 - Adding Gradle section to Getting Started docs and clarifying which repos are available and when snapshots etc should be used.
2013-02-09 13:00:41 +01:00
Björn Antonsson
a54659e5b4
We can't set the SHA1PRNG random source reliably from within Akka. See #2977
2013-02-08 16:20:40 +01:00
Patrik Nordwall
d32a2edc51
Buffer LeaderChanged events and publish all on convergence, see #3017
...
* Otherwise some changes might never be published, since it doesn't have
to be convergence on all nodes inbetween all transitions.
* Detected by a failure ClusterSingletonManagerSpec.
* Added a test to simulate the failure scenario.
2013-02-08 12:29:11 +01:00