[
https://issues.apache.org/jira/browse/MAPREDUCE-3851?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13209877#comment-13209877
]
Thomas Graves commented on MAPREDUCE-3851:
------------------------------------------
All the instances I've seen where it hits JETTY-1342 it is hosed pretty much
immediately and it doesn't recover after seeing just a couple of the
exceptions. That is why I had it just a straight count over the lifetime.
I guess it would make it more extensible for possible future Jetty bugs to have
it be a ratio though. So how about we do the ratio of say the last 100
requests? If that ratio goes above a certain threshold, then we abort. we
could make that number (100) configurable if someone really wants.
> Allow more aggressive action on detection of the jetty issue
> ------------------------------------------------------------
>
> Key: MAPREDUCE-3851
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-3851
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: tasktracker
> Affects Versions: 1.0.0
> Reporter: Kihwal Lee
> Assignee: Thomas Graves
> Fix For: 1.1.0, 1.0.1
>
> Attachments: MAPREDUCE-3851.patch
>
>
> MAPREDUCE-2529 added the useful failure detection mechanism. In this jira, I
> propose we add a periodic check inside TT and configurable action to
> self-destruct. Blacklisting helps but is not enough. Hung jetty still accepts
> connection and it takes very long time for clients to fail out. Short jobs
> are delayed for hours because of this. This feature will be a nice companion
> to MAPREDUCE-3184.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
