[jira] [Updated] (MAPREDUCE-5704) Optimize nextJobId in JobTracker

["Allen Wittenauer (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MAPREDUCE-5704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Allen Wittenauer updated MAPREDUCE-5704:
----------------------------------------
    Labels: BB2015-05-TBR  (was: )

> Optimize nextJobId in JobTracker
> --------------------------------
>
>                 Key: MAPREDUCE-5704
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-5704
>             Project: Hadoop Map/Reduce
>          Issue Type: Improvement
>          Components: jobtracker, mrv1
>    Affects Versions: 1.2.1
>            Reporter: JamesLi
>            Assignee: JamesLi
>              Labels: BB2015-05-TBR
>         Attachments: MAPREDUCE-5704.patch
>
>
> When jobtracker start, nextJobId start with 1,if we have run 3000 jobs  then restart jobtracker and run a new job,we can not see this new job on jobtracker:5030/jobhistory.jsp unless click "get more results" button.
> In jobhistory_jsp.java, array SCAN_SIZES controls job numbers displayed on jobhistory.jsp.
> I make a little chage,when jobtracker start,find the biggest id under history done directory,job will start with maxId+1 or 1 if can not find any job files.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)