[
https://issues.apache.org/jira/browse/BEAM-9021?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17007876#comment-17007876
]
Kyle Weaver commented on BEAM-9021:
-----------------------------------
Looks like job shutdown is hanging somewhere. Unhealthy stack trace:
10:10:12 [flink-akka.actor.default-dispatcher-3] INFO
org.apache.flink.runtime.resourcemanager.StandaloneResourceManager - Disconnect
job manager 9e7e6603d74418a618441f553ae2402f@akka://flink/user/jobmanager_1 for
job 2d6f99c94b4190d3c1b7518427a9c4ea from the resource manager.
10:10:12 [flink-akka.actor.default-dispatcher-7] INFO
org.apache.flink.runtime.taskexecutor.JobLeaderService - Remove job
2d6f99c94b4190d3c1b7518427a9c4ea from job leader monitoring.
10:10:12 [flink-akka.actor.default-dispatcher-7] INFO
org.apache.flink.runtime.taskexecutor.TaskExecutor - Close JobManager
connection for job 2d6f99c94b4190d3c1b7518427a9c4ea.
10:10:12 [flink-akka.actor.default-dispatcher-7] INFO
org.apache.flink.runtime.taskexecutor.TaskExecutor - Close JobManager
connection for job 2d6f99c94b4190d3c1b7518427a9c4ea.
10:10:12 [flink-akka.actor.default-dispatcher-7] INFO
org.apache.flink.runtime.taskexecutor.JobLeaderService - Cannot reconnect to
job 2d6f99c94b4190d3c1b7518427a9c4ea because it is not registered.
12:00:19 Build timed out (after 120 minutes). Marking the build as aborted.
versus healthy stack trace:
04:22:45 [flink-akka.actor.default-dispatcher-7] INFO
org.apache.flink.runtime.resourcemanager.StandaloneResourceManager - Disconnect
job manager 83397ba8115450ca96c19518ce584bfa@akka://flink/user/jobmanager_1 for
job a411f9f0385ea8c6c3bd6c35c4f4f0b7 from the resource manager.
04:22:45 [main] INFO org.apache.flink.runtime.minicluster.MiniCluster -
Shutting down Flink Mini Cluster
04:22:45 [main] INFO org.apache.flink.runtime.dispatcher.DispatcherRestEndpoint
- Shutting down rest endpoint.
04:22:45 [flink-akka.actor.default-dispatcher-2] INFO
org.apache.flink.runtime.taskexecutor.TaskExecutor - Close JobManager
connection for job a411f9f0385ea8c6c3bd6c35c4f4f0b7.
04:22:45 [flink-akka.actor.default-dispatcher-2] INFO
org.apache.flink.runtime.taskexecutor.JobLeaderService - Cannot reconnect to
job a411f9f0385ea8c6c3bd6c35c4f4f0b7 because it is not registered.
04:22:45 [flink-akka.actor.default-dispatcher-2] INFO
org.apache.flink.runtime.taskexecutor.TaskExecutor - Stopping TaskExecutor
akka://flink/user/taskmanager_0.
04:22:45 [flink-akka.actor.default-dispatcher-2] INFO
org.apache.flink.runtime.taskexecutor.TaskExecutor - Close ResourceManager
connection 128be3856c0fd9cd78735d7a83e5cc74.
> Flink portable runner postcommit times out
> ------------------------------------------
>
> Key: BEAM-9021
> URL: https://issues.apache.org/jira/browse/BEAM-9021
> Project: Beam
> Issue Type: Improvement
> Components: runner-flink, test-failures
> Reporter: Kyle Weaver
> Assignee: Kyle Weaver
> Priority: Major
> Labels: currently-failing, portability-flink
>
> Build times appear to be inconsistent, with some builds taking 10-30 mins and
> others timing out at 2 hours.
> https://builds.apache.org/job/beam_PostCommit_PortableJar_Flink/buildTimeTrend
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
