[
https://issues.apache.org/jira/browse/TWILL-80?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14245263#comment-14245263
]
Terence Yim commented on TWILL-80:
----------------------------------
The issue is in the {{KafkaAppender.forceFlush()}} call. When forceFlush is
called, it's possible that the changes in the concurrent LinkedList, which
suppose contains the latest log entry being emitted, may not be visible to the
forceFlush thread yet, hence the force flush is not flushing. The entries will
be left in the buffer and never get flushed when the appender shutdown.
> Logs not available when the runnable failed at initialize or only runs
> momentarily
> -----------------------------------------------------------------------------------
>
> Key: TWILL-80
> URL: https://issues.apache.org/jira/browse/TWILL-80
> Project: Apache Twill
> Issue Type: Bug
> Components: core
> Affects Versions: 0.2.0-incubating
> Reporter: Terence Yim
> Assignee: Terence Yim
> Fix For: 0.5.0-incubating
>
>
> Multiple users reported this issue. Need to investigate and fix.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
