hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jason Lowe (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (MAPREDUCE-4705) Historyserver links expire before the history data does
Date Mon, 08 Oct 2012 19:48:03 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-4705?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13471806#comment-13471806

Jason Lowe commented on MAPREDUCE-4705:

The problem occurs because HistoryFileManager.scanOldDirsForJob is using a job's serial number
directly to lookup the serial number index (which would require an entry in the map per job)
rather than the result of JobHistoryUtils.serialNumberDirectoryComponent which carves off
the lower digits of the serial number, effectively grouping them into buckets of 1000.  I'll
post a patch shortly.
> Historyserver links expire before the history data does
> -------------------------------------------------------
>                 Key: MAPREDUCE-4705
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4705
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: jobhistoryserver, mrv2
>    Affects Versions: 0.23.3
>            Reporter: Jason Lowe
>            Assignee: Jason Lowe
>            Priority: Critical
> The historyserver can serve up links to jobs that become useless well before the job
history files are purged.  For example on a large, heavily used cluster we can end up rotating
through the maximum number of jobs the historyserver can track fairly quickly.  If a user
was investigating an issue with a job using a saved historyserver URL, that URL can become
useless because the historyserver has forgotten about the job even though the history files
are still sitting in HDFS.
> We can tell the historyserver to keep track of more jobs by increasing {{mapreduce.jobhistory.joblist.cache.size}},
but this has a direct impact on the responsiveness of the main historyserver page since it
serves up all the entries to the client at once.  It looks like Hadoop 1.x avoided this issue
by encoding the history file location into the URLs served up by the historyserver, so it
didn't have to track a mapping between job ID and history file location.

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

View raw message