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

Posted to issues@spark.apache.org by "Apache Spark (Jira)" <ji...@apache.org> on 2021/07/20 17:59:00 UTC

[jira] [Commented] (SPARK-36215) Add logging for slow fetches to diagnose external shuffle service issues

    [ https://issues.apache.org/jira/browse/SPARK-36215?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17384428#comment-17384428 ] 

Apache Spark commented on SPARK-36215:
--------------------------------------

User 'shardulm94' has created a pull request for this issue:
https://github.com/apache/spark/pull/33446

> Add logging for slow fetches to diagnose external shuffle service issues
> ------------------------------------------------------------------------
>
>                 Key: SPARK-36215
>                 URL: https://issues.apache.org/jira/browse/SPARK-36215
>             Project: Spark
>          Issue Type: Improvement
>          Components: Shuffle
>    Affects Versions: 3.2.0
>            Reporter: Shardul Mahadik
>            Priority: Trivial
>
> Currently we can see from the metrics that a task or stage has slow fetches, and the logs indicate _all_ of the shuffle servers those tasks were fetching from, but often this is a big set (dozens or even hundreds) and narrowing down which one caused issues can be very difficult. We should add some logging when a fetch is "slow" as determined by some preconfigured thresholds.
>  



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

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