[
https://issues.apache.org/jira/browse/MAPREDUCE-6067?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14122473#comment-14122473
]
Todd Lipcon commented on MAPREDUCE-6067:
----------------------------------------
I tested Binglin's latest patch (v5) on the same cluster, and can confirm that
the counters are now correct. The total map task CPU time was about 0.9% worse,
but that's probably in the realm of noise (just did one run of each).
> native-task: spilled records counter is incorrect
> -------------------------------------------------
>
> Key: MAPREDUCE-6067
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-6067
> Project: Hadoop Map/Reduce
> Issue Type: Sub-task
> Components: task
> Reporter: Todd Lipcon
> Assignee: Binglin Chang
> Attachments: MAPREDUCE-6067.v1.patch, MAPREDUCE-6067.v2.patch,
> MAPREDUCE-6067.v3.patch, MAPREDUCE-6067.v4.patch, MAPREDUCE-6067.v5.patch,
> native-counters.html, trunk-counters.html
>
>
> After running a terasort, I see the spilled records counter at 5028651606,
> which is about half what I expected to see. Using the non-native collector I
> see the expected count of 10000000000. It seems the correct number of records
> were indeed spilled, because the job's output record count is correct.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
