[
https://issues.apache.org/jira/browse/MAPREDUCE-5533?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vinod Kumar Vavilapalli updated MAPREDUCE-5533:
-----------------------------------------------
Status: Open (was: Patch Available)
I presume this is not just for reduces. Even maps that are 'paused' will run
into the same issue? If so, please edit the ticket title.
Also we already have code to handle this. Though estimatedRunTime doesn't
change for paused task-attempts, we look at estimatedEndTime,
estimatedReplacementEndTime and speculate if needed. That code should just work
fine even in this case, although it is clear there is some bug somewhere. This
isn't the correct fix, let's investigate more.
> Speculative execution does not function for reduce
> --------------------------------------------------
>
> Key: MAPREDUCE-5533
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-5533
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: applicationmaster
> Reporter: Tassapol Athiapinya
> Assignee: Xuan Gong
> Attachments: MAPREDUCE-5533.1.patch, MAPREDUCE-5533.2.patch,
> MAPREDUCE-5533.3.patch, MAPREDUCE-5533.4.patch
>
>
> We have sort job where reduce attempt does not send heartbeat in timely
> manner to application master. The AM should kick off another attempt to let
> job succeeds. What we find is the job fails and there is no speculation
> happening.
--
This message was sent by Atlassian JIRA
(v6.1#6144)