You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@ozone.apache.org by "Arpit Agarwal (Jira)" <ji...@apache.org> on 2023/08/16 15:06:00 UTC

[jira] [Resolved] (HDDS-782) MR example pi job runs 5 min for 1 Map/1 Sample

     [ https://issues.apache.org/jira/browse/HDDS-782?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Arpit Agarwal resolved HDDS-782.
--------------------------------
    Resolution: Abandoned

Let's abandon this one, we have had a huge number of performance improvements in the last 4-5 years, and this is likely obsolete. We can reopen if we still suspect a different perf issue.

> MR example pi job runs 5 min for 1 Map/1 Sample
> -----------------------------------------------
>
>                 Key: HDDS-782
>                 URL: https://issues.apache.org/jira/browse/HDDS-782
>             Project: Apache Ozone
>          Issue Type: Bug
>          Components: Ozone Filesystem
>    Affects Versions: 0.3.0
>            Reporter: Soumitra Sulav
>            Priority: Major
>         Attachments: ozoneMRJob.log
>
>
> On running a hadoop examples pi job it takes 250+ seconds which generally run in few seconds in HDFS cluster.
> On seeing the service/job logs it seems that there are few _SocketTimeoutException_ in between and YARN keeps on _Waiting for AsyncDispatcher to drain_. Thread state is _WAITING_ for a very long interval.
>  
> Refer attached log for further details.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

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