[
https://issues.apache.org/jira/browse/SAMZA-414?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Chris Riccomini updated SAMZA-414:
----------------------------------
Attachment: log-hprof.png
SAMZA-414-1.patch
Attaching updated patch.
I switched the hprof dump location from $base_dir/tmp to $SAMZA_LOG_DIR. The
reason for this change is that, by default, YARN deletes the $base_dir
immediately after a container finishes. This behavior can be tuned with
[yarn.nodemanager.delete.debug-delay-sec|http://hadoop.apache.org/docs/r2.4.0/hadoop-yarn/hadoop-yarn-common/yarn-default.xml],
but no one does by default.
In an effort to make things easier, and do things right in the default
scenario, the log directory seemed appropriate. I'm attaching a screenshot that
shows how it works. Since the hprof file is binary, you can't actually view it
in the web UI, but you can at least find it, and SSH to the proper box to get
at it.
> Enable HeapDumpOnOutOfMemoryError by default
> --------------------------------------------
>
> Key: SAMZA-414
> URL: https://issues.apache.org/jira/browse/SAMZA-414
> Project: Samza
> Issue Type: Bug
> Components: container
> Affects Versions: 0.8.0
> Reporter: Chris Riccomini
> Fix For: 0.8.0
>
> Attachments: SAMZA-414-0.patch, SAMZA-414-1.patch, log-hprof.png
>
>
> It would be nice if Samza's run-class.sh defaulted to use
> -XX:+HeapDumpOnOutOfMemoryError. According to
> [this|http://stackoverflow.com/questions/542979/using-heapdumponoutofmemoryerror-parameter-for-heap-dump-for-jboss]
> post, it puts the heap dump in CWD by default, which should be fine.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
