You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by rmetzger <gi...@git.apache.org> on 2015/11/24 16:51:29 UTC
[GitHub] flink pull request: [hotfix] Register signal handler for JobManage...
GitHub user rmetzger opened a pull request:
https://github.com/apache/flink/pull/1400
[hotfix] Register signal handler for JobManager and TaskManager
I've spend some time today helping a user to debug an issue with a TaskManager which was stopping for no apparent reason.
The following changes in the code will make it easier in the future to resolve similar issues.
The BlobCache (running on the TM) will log that its shutting down.
Also, the JM and TM will log when they get killed by an external system.
I didn't push this hotfix immediately because I might have overseen an entry point into the JM / TM.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/rmetzger/flink debug_hotfix
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/flink/pull/1400.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 #1400
----
commit 3dff2ba436406782043ed24d1a7adeb1d5998276
Author: Robert Metzger <rm...@apache.org>
Date: 2015-11-24T15:49:35Z
[hotfix] Register signal handler for JobManager and TaskManager
----
---
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: [hotfix] Register signal handler for JobManage...
Posted by mxm <gi...@git.apache.org>.
Github user mxm commented on the pull request:
https://github.com/apache/flink/pull/1400#issuecomment-161350648
I think the reporting is a bit aggressive. I wouldn't throw an ERROR on a regular shutdown of a cluster.
```
16:03:22,877 ERROR org.apache.flink.runtime.jobmanager.JobManager - RECEIVED SIGNAL 15: SIGTERM
16:03:22,878 ERROR org.apache.flink.runtime.jobmanager.JobManager - This JVM will shut down because it was killed from the outside.
```
---
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: [hotfix] Register signal handler for JobManage...
Posted by asfgit <gi...@git.apache.org>.
Github user asfgit closed the pull request at:
https://github.com/apache/flink/pull/1400
---
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: [hotfix] Register signal handler for JobManage...
Posted by rmetzger <gi...@git.apache.org>.
Github user rmetzger commented on the pull request:
https://github.com/apache/flink/pull/1400#issuecomment-159638066
Great idea. Will do & merge.
---
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: [hotfix] Register signal handler for JobManage...
Posted by StephanEwen <gi...@git.apache.org>.
Github user StephanEwen commented on the pull request:
https://github.com/apache/flink/pull/1400#issuecomment-159566115
Can we change the signal handler such that it prints on "TERM" and "KILL" a message that explicitly says that the process was killed from the outside? This should save us from many clarification questions on the mailing list ;-)
---
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.
---