[
https://issues.apache.org/jira/browse/MAPREDUCE-3921?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13261127#comment-13261127
]
Bikas Saha commented on MAPREDUCE-3921:
---------------------------------------
I see what you are saying about fetch failures and bad nodes. I am open to both
approaches. The way its currently done is based off discussions I had with
Vinod long ago.
Changed JobImpl to ignore JOB_UPDATED_NODE events in NEW and INITED states.
Changed to TaskId.getTaskType()
Yes, the removed entry is marked OBSOLETE. But I still dont understand why that
would be done if the current entry is not successful. Why lose the previously
successful entry when the current one is not successful itself? This should be
done only when the current entry is successful.
I like the change to store NodeId's instead of ContainerId's in the
AssignedRequests map. I would like to make it a separate change and not merge
it with this one. There might be other gotchas to doing that.
> MR AM should act on the nodes liveliness information when nodes go
> up/down/unhealthy
> ------------------------------------------------------------------------------------
>
> Key: MAPREDUCE-3921
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-3921
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: mr-am, mrv2
> Affects Versions: 0.23.0
> Reporter: Vinod Kumar Vavilapalli
> Assignee: Bikas Saha
> Fix For: 0.23.2
>
> Attachments: MAPREDUCE-3921-1.patch, MAPREDUCE-3921-3.patch,
> MAPREDUCE-3921-4.patch, MAPREDUCE-3921-5.patch,
> MAPREDUCE-3921-branch-0.23.patch, MAPREDUCE-3921-branch-0.23.patch,
> MAPREDUCE-3921-branch-0.23.patch, MAPREDUCE-3921.patch
>
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
