hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Todd Lipcon (JIRA)" <j...@apache.org>
Subject [jira] Updated: (MAPREDUCE-2051) Contribute a fair scheduler preemption system test
Date Mon, 06 Sep 2010 22:45:33 GMT

     [ https://issues.apache.org/jira/browse/MAPREDUCE-2051?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Todd Lipcon updated MAPREDUCE-2051:

    Status: Patch Available  (was: Open)

Oops, forgot to hit Submit Patch. This test is nothing fancy, but when run with jcarder it
at least exercises the "real" code paths so will pick up inconsistent lock orderings which
often creep in in the JT with respect to the scheduler (JT, Scheduler, JIP, and TIP locks
in particular)

> Contribute a fair scheduler preemption system test
> --------------------------------------------------
>                 Key: MAPREDUCE-2051
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-2051
>             Project: Hadoop Map/Reduce
>          Issue Type: Test
>          Components: contrib/fair-share
>    Affects Versions: 0.22.0
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>         Attachments: mapreduce-2051.txt
> We've seen a number of bugs in the fair share scheduler not caught by its unit tests,
which are heavily mock-based. This JIRA is to add an integration/system/stress test for the
fairshare scheduler. This test can help identify races and deadlocks, and when run within
the jcarder framework has identified several potential deadlocks for us that aren't turned
up running small scale testing.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message