[
https://issues.apache.org/jira/browse/LOG4J2-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15358087#comment-15358087
]
Remko Popma commented on LOG4J2-1453:
-------------------------------------
A long to represent timeout duration in millis is fine (and would be consistent
with the way [AsyncAppender
timeout|https://logging.apache.org/log4j/2.x/manual/appenders.html#AsyncAppender]
is configured).
> Configurable shutdown timeout for AsyncLogger
> ---------------------------------------------
>
> Key: LOG4J2-1453
> URL: https://issues.apache.org/jira/browse/LOG4J2-1453
> Project: Log4j 2
> Issue Type: Improvement
> Components: Core
> Affects Versions: 2.6.1
> Reporter: David Leonhartsberger
> Priority: Minor
> Original Estimate: 28h
> Remaining Estimate: 28h
>
> The current implementation of {code}AsyncLoggerDisruptor#stop(){code} could
> block for a long time waiting for the Disruptor queue to be processed.
> I suggest to use a configurable timeout (similar to AsyncAppender) and call
> the {code}Disruptor#shutdown(final long timeout, final TimeUnit
> timeUnit){code} method instead of the one which takes no paramter and
> therefore waits indefinitely.
> If there are still unconsumed log entries left in the Disruptor after the
> timeout, I would just log the count of discarded entries as its done in
> AsyncAppender.stop.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
