[
https://issues.apache.org/jira/browse/MAPREDUCE-2386?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14053103#comment-14053103
]
Dmitry Sivachenko commented on MAPREDUCE-2386:
----------------------------------------------
I see the same errors with hadoop 2.4.1 on FreeBSD:
2014-07-06 16:06:27,063 INFO [1707749244@qtp-1646583765-1 - Acceptor0
selectchannelconnec...@nezabudka3-00.yandex.ru:8088] mortbay.log
(Slf4jLog.java:info(67)) -
org.mortbay.io.nio.SelectorManager$SelectSet@2e0c0189 JVM BUG(s) - injecting
delay57 times
2014-07-06 16:06:27,063 INFO [1707749244@qtp-1646583765-1 - Acceptor0
selectchannelconnec...@nezabudka3-00.yandex.ru:8088] mortbay.log
(Slf4jLog.java:info(67)) -
org.mortbay.io.nio.SelectorManager$SelectSet@2e0c0189 JVM BUG(s) - recreating
selector 57 times, canceled keys 915 times
2014-07-06 16:07:27,064 INFO [1707749244@qtp-1646583765-1 - Acceptor0
selectchannelconnec...@nezabudka3-00.yandex.ru:8088] mortbay.log
(Slf4jLog.java:info(67)) -
org.mortbay.io.nio.SelectorManager$SelectSet@2e0c0189 JVM BUG(s) - injecting
delay58 times
Why Jetty can't be updated to at least version 7
(https://issues.apache.org/jira/browse/HADOOP-8515 has a patch for more that a
year)?
> TT jetty server stuck in tight loop around epoll_wait
> -----------------------------------------------------
>
> Key: MAPREDUCE-2386
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-2386
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: tasktracker
> Affects Versions: 0.23.0, 1.0.3
> Environment: RHEL 6.0 "Santiago"
> Reporter: Todd Lipcon
>
> In some load testing, I got a TaskTracker into a state where its Jetty server
> is in a tight loop calling epoll_wait, which is returning EINVAL:
> [pid 19573] epoll_wait(157, 40829000, 8192, 0) = -1 EINVAL (Invalid argument)
> It's not responding to any HTTP connections - connections are accepted and
> then just hang.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
