[
https://issues.apache.org/jira/browse/FLUME-1601?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13460647#comment-13460647
]
Brock Noland commented on FLUME-1601:
-------------------------------------
I guess what I am saying is that we cannot guarantee the implementations of
MappedByteBuffer and LogFile.MetaData writers will never do something
interruptible like use channel io. It'd be very easy to introduce an
interruptible io into the backing store checkpoint or metadata updates and it's
almost impossible to test. We can achieve the same affect as we have now in a
more safe fashion by using a blocking queue.
> FileChannel log.close calls interrupt on the background worker which could
> corrupt checkpoint
> ---------------------------------------------------------------------------------------------
>
> Key: FLUME-1601
> URL: https://issues.apache.org/jira/browse/FLUME-1601
> Project: Flume
> Issue Type: Improvement
> Components: Channel
> Affects Versions: v1.2.0
> Reporter: Brock Noland
> Assignee: Brock Noland
> Priority: Minor
>
> Since the background worker is writing out the checkpoint, we should not
> interrupt it.
--
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
