Mailing-List: contact mapreduce-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Tue, 7 Mar 2017 19:14:38 +0000 (UTC)
From: "Yufei Gu (JIRA)" <jira@apache.org>
To: mapreduce-issues@hadoop.apache.org
Message-ID: <JIRA.13048985.1488913978000.4329.1488914078238@Atlassian.JIRA>
In-Reply-To: <JIRA.13048985.1488913978000@Atlassian.JIRA>
References: <JIRA.13048985.1488913978000@Atlassian.JIRA> <JIRA.13048985.1488913978665@jira-lw-us.apache.org>
Subject: [jira] [Updated] (MAPREDUCE-6858) HistoryFileManager thrashing due
 to high volume jobs
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Tue, 07 Mar 2017 19:14:43 -0000


     [ https://issues.apache.org/jira/browse/MAPREDUCE-6858?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Yufei Gu updated MAPREDUCE-6858:
--------------------------------
    Description: 
- JHS scans "done_intermediate" dir for files to process and adds them to a thread pool
- Thread pool starts processing these files to move them to "done" dir
- JHS scans "done_intermediate" again for files to process and adds them to a thread pool
-- If we have enough jobs where the thread pool can't keep up with the scanning interval, they'll get added twice (or more). If this keeps compounding,  jobs end up would pile up and not getting processed for quite some time and getting lots of FileNotFoundException's.

By default, it looks like the thread pool only has 3 threads in it (mapreduce.jobhistory.move.thread-count). And the scan interval is 3 minutes (mapreduce.jobhistory.move.interval-ms). Perhaps we should increase these?

  was:
- JHS scans "done_intermediate" dir for files to process and adds them to a thread pool
- Thread pool starts processing these files to move them to "done" dir
- JHS scans "done_intermediate" again for files to process and adds them to a thread pool
-- If we have enough jobs where the thread pool can't keep up with the scanning interval, they'll get added twice (or more). If this keeps compounding, I wouldn't be surprised if jobs end up piling up and not getting processed for quite some time and getting lots of FileNotFoundException's.

By default, it looks like the thread pool only has 3 threads in it (mapreduce.jobhistory.move.thread-count). And the scan interval is 3 minutes (mapreduce.jobhistory.move.interval-ms). Perhaps we should increase these?


> HistoryFileManager thrashing due to high volume jobs 
> -----------------------------------------------------
>
>                 Key: MAPREDUCE-6858
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-6858
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: jobhistoryserver
>            Reporter: Yufei Gu
>
> - JHS scans "done_intermediate" dir for files to process and adds them to a thread pool
> - Thread pool starts processing these files to move them to "done" dir
> - JHS scans "done_intermediate" again for files to process and adds them to a thread pool
> -- If we have enough jobs where the thread pool can't keep up with the scanning interval, they'll get added twice (or more). If this keeps compounding,  jobs end up would pile up and not getting processed for quite some time and getting lots of FileNotFoundException's.
> By default, it looks like the thread pool only has 3 threads in it (mapreduce.jobhistory.move.thread-count). And the scan interval is 3 minutes (mapreduce.jobhistory.move.interval-ms). Perhaps we should increase these?


--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

---------------------------------------------------------------------
To unsubscribe, e-mail: mapreduce-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: mapreduce-issues-help@hadoop.apache.org