[
https://issues.apache.org/jira/browse/MAPREDUCE-6975?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16198203#comment-16198203
]
Prabhu Joseph commented on MAPREDUCE-6975:
------------------------------------------
[~Naganarasimha] Will add the output as a comment in code. Have tried with
Counters.toSting() which outputs each counter in a separate line. I think
having all counters in same line along with task id will be easier to compare
the counters of each task with awk and grep. We can make it readable by not
printing the group name for every counter the Group has which will reduce the
verbosity. Shall we refactor the Counters.makeCompactString similar to below
way what Tez uses.
https://github.com/apache/tez/blob/master/tez-api/src/main/java/org/apache/tez/common/counters/AbstractCounters.java#L314
{code}
2017-10-05 16:00:40,542 INFO [main] org.apache.hadoop.mapred.Task: Final
Counters for attempt_1506610341926_0006_m_000000_0: [[<groupname1>
countername=value, countername1=value1,][<groupname2> countername=value,
countername1=value1,]]
{code}
> Logging task counters
> ----------------------
>
> Key: MAPREDUCE-6975
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-6975
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: task
> Affects Versions: 2.7.3
> Reporter: Prabhu Joseph
> Assignee: Prabhu Joseph
> Attachments: MAPREDUCE-6975.1.patch, MAPREDUCE-6975.patch
>
>
> Logging counters for each task at the end of it's syslog will make debug
> easier with just application logs.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
