hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Arpit Gupta (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-1097) Yarn client reports the job as failed when RM restarts while the job is still running and eventually finishes
Date Mon, 26 Aug 2013 18:07:52 GMT

    [ https://issues.apache.org/jira/browse/YARN-1097?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13750337#comment-13750337
] 

Arpit Gupta commented on YARN-1097:
-----------------------------------

For example here is a small snippet of the client

{code}
2013-08-26 07:37:28,147 INFO  [main] mapreduce.Job:  map 76% reduce 5%
2013-08-26 07:37:29,152 INFO  [main] mapreduce.Job:  map 77% reduce 6%
2013-08-26 07:37:31,160 INFO  [main] mapreduce.Job:  map 78% reduce 6%
2013-08-26 07:37:34,181 INFO  [main] mapreduce.Job:  map 79% reduce 6%
2013-08-26 07:37:39,205 INFO  [main] mapreduce.Job:  map 79% reduce 7%
2013-08-26 07:37:40,210 INFO  [main] mapreduce.Job:  map 80% reduce 7%
2013-08-26 07:37:40,215 INFO  [main] mapreduce.Job:  map 80% reduce 1%
2013-08-26 07:37:40,219 INFO  [main] mapreduce.Job: Job job_1377500778739_0003 failed with
state FAILED due to:
2013-08-26 07:37:40,294 INFO  [main] mapreduce.Job: Counters: 0
{code}

For this test the RM was killed at 7:37:09 and started back up at 7:37:19
                
> Yarn client reports the job as failed when RM restarts while the job is still running
and eventually finishes
> -------------------------------------------------------------------------------------------------------------
>
>                 Key: YARN-1097
>                 URL: https://issues.apache.org/jira/browse/YARN-1097
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: resourcemanager
>    Affects Versions: 2.1.0-beta
>            Reporter: Arpit Gupta
>            Assignee: Vinod Kumar Vavilapalli
>
> While running rm restart tests it was noticed that the yarn client will some times prematurely
report a job as failed where as the job actually finishes to completion.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message