[jira] Commented: (MAPREDUCE-1043) JobHistory.parseHistoryFromFS does not appear to get us the information in conf files that correspond to jobs.

["Amar Kamat (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/MAPREDUCE-1043?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12771865#action_12771865 ] 

Amar Kamat commented on MAPREDUCE-1043:
---------------------------------------

Assuming MAPREDUCE-173 is fixed, how difficult is to do something like 
{code}
JobConf conf = new JobConf(value returned by JOBCONF key);
String value = conf.get(key);
{code}
Am I missing something?

> JobHistory.parseHistoryFromFS does not appear to get us the information in conf files that correspond to jobs.
> --------------------------------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-1043
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-1043
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>            Reporter: Dick King
>            Priority: Minor
>
> We get to see the NAMES of the files in the Listener callbacks, but that doesn't seem quite right because the user is left to parse the conf file hirself.  Perhaps there should be an additional interface that takes a Listener callback string that came with a JOBCONF Key, and delivers an abstraction instance that can be called with its own Listener that in turn will be called with each property/value pair from the named conf file?
> rumen can be changed to use JobHistory, but it will need to deal with this situation in an ad hoc way unless we make this change.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.