You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@kafka.apache.org by "ASF GitHub Bot (JIRA)" <ji...@apache.org> on 2018/05/23 11:43:00 UTC
[jira] [Commented] (KAFKA-4290) High CPU caused by timeout overflow
in WorkerCoordinator
[ https://issues.apache.org/jira/browse/KAFKA-4290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16487117#comment-16487117 ]
ASF GitHub Bot commented on KAFKA-4290:
---------------------------------------
GitHub user thisthat opened a pull request:
https://github.com/apache/activemq/pull/284
Avoid overflow errors with timestamps
Some comparisons with timestamp values are not safe. This comparisons can trigger errors that were found in other Apache projects, e.g. KAFKA-4290.
I changed those comparisons according to what the Java documentation recommends to help preventing such errors.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/thisthat/activemq following_java_rec
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/activemq/pull/284.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #284
----
commit 92a1926aa52de7d99142458ee74a6bb99cc5a963
Author: giliva <gi...@...>
Date: 2018-05-23T11:35:38Z
Avoid overflow errors - see KAFKA-4290
----
> High CPU caused by timeout overflow in WorkerCoordinator
> --------------------------------------------------------
>
> Key: KAFKA-4290
> URL: https://issues.apache.org/jira/browse/KAFKA-4290
> Project: Kafka
> Issue Type: Bug
> Components: KafkaConnect
> Reporter: Jason Gustafson
> Assignee: Jason Gustafson
> Priority: Blocker
> Fix For: 0.10.1.0, 0.10.2.0
>
>
> The timeout passed to {{WorkerCoordinator.poll()}} can overflow if large enough because we add it to the current time in order to calculate the call's deadline. This shortcuts the poll loop and results in a very tight event loop which can saturate a CPU. We hit this case out of the box because Connect uses a default timeout of {{Long.MAX_VALUE}}.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)