You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by tillrohrmann <gi...@git.apache.org> on 2016/10/27 09:46:26 UTC

[GitHub] flink pull request #2700: [FLINK-4933] [exec graph] Don't let the EG fail in...

GitHub user tillrohrmann opened a pull request:

    https://github.com/apache/flink/pull/2700

    [FLINK-4933] [exec graph] Don't let the EG fail in case of a failing scheduleOrUpdateConsumers call

    Instead of failing the complete ExecutionGraph, a failing scheduleOrUpdateConsumers call
    will be reported back to the caller. The caller can then decide what to do. Per default,
    it will fail the calling task.

You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/tillrohrmann/flink fixScheduleOrUpdateConsumers

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/flink/pull/2700.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #2700
    
----
commit e51f0a56762c7f12acc215a53ffce2af28d38583
Author: Till Rohrmann <tr...@apache.org>
Date:   2016-10-27T09:41:29Z

    [FLINK-4933] [exec graph] Don't let the EG fail in case of a failing scheduleOrUpdateConsumers call
    
    Instead of failing the complete ExecutionGraph, a failing scheduleOrUpdateConsumers call
    will be reported back to the caller. The caller can then decide what to do. Per default,
    it will fail the calling task.

----


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] flink pull request #2700: [FLINK-4933] [exec graph] Don't let the EG fail in...

Posted by uce <gi...@git.apache.org>.
Github user uce commented on a diff in the pull request:

    https://github.com/apache/flink/pull/2700#discussion_r85533357
  
    --- Diff: flink-runtime/src/main/scala/org/apache/flink/runtime/jobmanager/JobManager.scala ---
    @@ -917,8 +917,15 @@ class JobManager(
         case ScheduleOrUpdateConsumers(jobId, partitionId) =>
           currentJobs.get(jobId) match {
             case Some((executionGraph, _)) =>
    -          sender ! decorateMessage(Acknowledge)
    -          executionGraph.scheduleOrUpdateConsumers(partitionId)
    +          try {
    +            executionGraph.scheduleOrUpdateConsumers(partitionId)
    +            sender ! decorateMessage(Acknowledge)
    +          } catch {
    +            case e: ExecutionGraphException =>
    --- End diff --
    
    Does it make sense to catch the more generic `Exception` type here in order to make the sender notice any problems sooner? I see that the method only throws EGExceptions currently but maybe at some point in time someone introduces a runtime exception etc. This would only be logged at the JM and the task's ask would timeout.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] flink pull request #2700: [FLINK-4933] [exec graph] Don't let the EG fail in...

Posted by asfgit <gi...@git.apache.org>.
Github user asfgit closed the pull request at:

    https://github.com/apache/flink/pull/2700


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] flink issue #2700: [FLINK-4933] [exec graph] Don't let the EG fail in case o...

Posted by StephanEwen <gi...@git.apache.org>.
Github user StephanEwen commented on the issue:

    https://github.com/apache/flink/pull/2700
  
    Looks good to me, +1
    
    Merging this...


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

[GitHub] flink issue #2700: [FLINK-4933] [exec graph] Don't let the EG fail in case o...

Posted by tillrohrmann <gi...@git.apache.org>.
Github user tillrohrmann commented on the issue:

    https://github.com/apache/flink/pull/2700
  
    Thanks for the review @uce. This is a problem which is relevant for the current master. It could make the `ExecutionGraph` go into state `FAILED` when being in state `RESTARTING` in case that the `scheduleOrUpdateConsumers` call failed. 
    
    Will rebase the PR because it was based on the failing `SpanningRecordSerializerTest`.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---