[
https://issues.apache.org/jira/browse/GEODE-8873?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17273941#comment-17273941
]
ASF GitHub Bot commented on GEODE-8873:
---------------------------------------
DonalEvans commented on pull request #5969:
URL: https://github.com/apache/geode/pull/5969#issuecomment-769301760
> This fixes the problem but I'm wondering if the code would be cleaner if
you did just one null check on _list_ like
>
> > if (list == null) {
> > list = Collections.emptyList();
> > }
>
> Then you could get rid of the rest of the null checks.
I took a slightly different approach to the one suggested, since if `peek()`
throws an exception, it would still have been possible for `list` to be null in
the last two places it was used. Instantiating `list` at the time of
declaration and then using `addAll()` prevents it from ever being null.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> MessageDispatcher logs Exception stack trace when stopped
> ---------------------------------------------------------
>
> Key: GEODE-8873
> URL: https://issues.apache.org/jira/browse/GEODE-8873
> Project: Geode
> Issue Type: Improvement
> Components: logging
> Affects Versions: 1.12.0, 1.13.1, 1.14.0
> Reporter: Donal Evans
> Assignee: Donal Evans
> Priority: Major
> Labels: pull-request-available
>
> This message is logged when there are still messages in the queue when the
> cache server thread receives the stop signal or client leaving signal and
> tries to clean up the messages in the queue but encounters an exception. In
> the case of a {{RegionDestroyedException}} orĀ "Broken Pipe" {{IOException}},
> a brief explanation of the cause is logged, but for any other exception
> encountered when attempting to send messages from the queue results in a full
> stack trace being logged, even though the log level is only {{info}}.
> The logging should be changed to only log the exception message, not a full
> stack trace.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)