[
https://issues.apache.org/jira/browse/YARN-7244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16181173#comment-16181173
]
Sunil G commented on YARN-7244:
-------------------------------
Thanks [~kshukla] for this improvement.
Some doubts here.
# ShuffleHandler is in mapred package. I think its not that safe to read
YARN_SHUFFLE_BAD_DIRS_FILTER_ENABLED from MR end. We could push this config
name to LocalDirAllocator and then read from NM end. Please correct me if I
missed something.
# A pull model to get all changed dirs from NM needs context as you mentioned.
{{lDirAlloc.getLocalPathToRead}} could definitely skip failed/bad dirs on fly.
If all dirs are bad, then MR has to be notified via exception. Do you think, we
can improve this to skip as default behavior itself?
> ShuffleHandler is not aware of disks that are added
> ---------------------------------------------------
>
> Key: YARN-7244
> URL: https://issues.apache.org/jira/browse/YARN-7244
> Project: Hadoop YARN
> Issue Type: Bug
> Reporter: Kuhu Shukla
> Assignee: Kuhu Shukla
> Attachments: YARN-7244.001.patch, YARN-7244.002.patch
>
>
> The ShuffleHandler permanently remembers the list of "good" disks on NM
> startup. If disks later are added to the node then map tasks will start using
> them but the ShuffleHandler will not be aware of them. The end result is that
> the data cannot be shuffled from the node leading to fetch failures and
> re-runs of the map tasks.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
