[
https://issues.apache.org/jira/browse/FLINK-2097?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14736438#comment-14736438
]
ASF GitHub Bot commented on FLINK-2097:
---------------------------------------
Github user mxm commented on a diff in the pull request:
https://github.com/apache/flink/pull/858#discussion_r39017774
--- Diff:
flink-tests/src/test/scala/org/apache/flink/api/scala/runtime/jobmanager/JobManagerFailsITCase.scala
---
@@ -53,41 +54,41 @@ class JobManagerFailsITCase(_system: ActorSystem)
}
"A TaskManager" should {
- "detect a lost connection to the JobManager and try to reconnect to
it" in {
-
- val num_slots = 13
- val cluster = startDeathwatchCluster(num_slots, 1)
-
- val tm = cluster.getTaskManagers(0)
- val jmGateway =
cluster.getLeaderGateway(TestingUtils.TESTING_DURATION)
-
- // disable disconnect message to test death watch
- tm ! DisableDisconnect
-
- try {
- within(TestingUtils.TESTING_DURATION) {
- jmGateway.tell(RequestNumberRegisteredTaskManager, self)
- expectMsg(1)
-
- tm ! NotifyWhenJobManagerTerminated(jmGateway.actor)
-
- jmGateway.tell(PoisonPill, self)
-
- expectMsgClass(classOf[JobManagerTerminated])
-
- cluster.restartLeadingJobManager()
-
- cluster.waitForTaskManagersToBeRegistered()
-
- cluster.getLeaderGateway(TestingUtils.TESTING_DURATION)
- .tell(RequestNumberRegisteredTaskManager, self)
-
- expectMsg(1)
- }
- } finally {
- cluster.stop()
- }
- }
+// "detect a lost connection to the JobManager and try to reconnect to
it" in {
--- End diff --
Thanks.
> Add support for JobSessions
> ---------------------------
>
> Key: FLINK-2097
> URL: https://issues.apache.org/jira/browse/FLINK-2097
> Project: Flink
> Issue Type: Sub-task
> Components: JobManager
> Affects Versions: 0.9
> Reporter: Stephan Ewen
> Assignee: Maximilian Michels
> Fix For: 0.10
>
>
> Sessions make sure that the JobManager does not immediately discard a
> JobGraph after execution, but keeps it around for further operations to be
> attached to the graph. By keeping the JobGraph around, the cached streams
> (intermediate data) are also kept,
> That is the way of realizing interactive sessions on top of a streaming
> dataflow abstraction.
> ExecutionGraphs should be kept as long as
> - no timeout occurred or
> - the session has not been explicitly ended
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
