[
https://issues.apache.org/jira/browse/SAMZA-14?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13740699#comment-13740699
]
Chris Riccomini commented on SAMZA-14:
--------------------------------------
logging stacktrace: Thinking more on this. I think the real issue here is that
we're treating two different kinds of errors the same way. There are
recoverable exceptions, like what I'm describing, and exceptions that will
never recover, like what you're describing. I think it's best to catch, log,
and retry my way for recoverable exceptions, and die with a full stack trace
for your kind. The easiest way to do that is to simply not catch the exception.
I think we just need to move the start* methods in SamzaContainer inside the
try/catch.
As for which exceptions to catch, I think we'll have to experiment. There's
probably some subset of kafka-related exceptions that we should retry. The rest
we can probably fail.
> Fix misc bugs in QA jobs
> ------------------------
>
> Key: SAMZA-14
> URL: https://issues.apache.org/jira/browse/SAMZA-14
> Project: Samza
> Issue Type: Bug
> Reporter: Jay Kreps
> Assignee: Jay Kreps
> Attachments: SAMZA-14-v1.patch
>
>
> Some bugs that have already been there and some that came with the various
> package refactorings.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
