You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@oozie.apache.org by "Azrael (JIRA)" <ji...@apache.org> on 2014/08/20 04:20:19 UTC

[jira] [Commented] (OOZIE-1867) specifying wrong jobtracker port hangs oozie job in setup process

    [ https://issues.apache.org/jira/browse/OOZIE-1867?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14103272#comment-14103272 ] 

Azrael commented on OOZIE-1867:
-------------------------------

Currently(trunk), It throws error with TRANSIENT type, so try 3 times and finally SUSPENDED.

Here is example job.properies:
{{jobTracker}} is set to {{localhost:9002}} instead of {{localhost:9001}}
{code}
ACTION[0000002-140819162858310-oozie-seoe-W@shell-1] Creating Job Client for action 0000002-140819162858310-oozie-seoe-W@shell-1
2014-08-19 17:58:40,062  WARN ActionStartXCommand:522 [pool-2-thread-1] - USER[seoeun] GROUP[-] TOKEN[] APP[test] JOB[0000002-140819162858310-oozie-seoe-W] ACTION[0000002-140819162858310-oozie-seoe-W@shell-1] Error starting action [shell-1]. ErrorType [TRANSIENT], ErrorCode [  JA006], Message [  JA006: Call to localhost/127.0.0.1:9002 failed on connection exception: java.net.ConnectException: Connection refused]
org.apache.oozie.action.ActionExecutorException:   JA006: Call to localhost/127.0.0.1:9002 failed on connection exception: java.net.ConnectException: Connection refused
{code}






> specifying wrong jobtracker port hangs oozie job in setup process
> -----------------------------------------------------------------
>
>                 Key: OOZIE-1867
>                 URL: https://issues.apache.org/jira/browse/OOZIE-1867
>             Project: Oozie
>          Issue Type: Improvement
>          Components: workflow
>    Affects Versions: 3.3.2
>         Environment: Hortonworks HDP 1.3
>            Reporter: Manish Nema
>
> When specify any working port which is not default jobtracker IPC port. Oozie submits workflow and remain in setup at MapReduce state forever. Instead it should time out and tell the unable to submit the job at the cluster.



--
This message was sent by Atlassian JIRA
(v6.2#6252)