[
https://issues.apache.org/jira/browse/KAFKA-1544?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14064989#comment-14064989
]
Jun Rao commented on KAFKA-1544:
--------------------------------
One solution is to simply put the sleep in a loop. In each iteration, we sleep
for a small amount of time, say 500ms. The loop finishes if the thread is not
runnable or if the backoff time is reached.
> LogCleaner may take a long time to shutdown
> -------------------------------------------
>
> Key: KAFKA-1544
> URL: https://issues.apache.org/jira/browse/KAFKA-1544
> Project: Kafka
> Issue Type: Bug
> Components: core
> Affects Versions: 0.8.2
> Reporter: Jun Rao
> Labels: newbie
>
> We have the following code in LogCleaner. Since the cleaner thread is
> shutdown w/o interrupt. If may take up to backoff time for the cleaner thread
> to detect the shutdown flag.
> private def cleanOrSleep() {
> cleanerManager.grabFilthiestLog() match {
> case None =>
> // there are no cleanable logs, sleep a while
> time.sleep(config.backOffMs)
>
--
This message was sent by Atlassian JIRA
(v6.2#6252)
