[
https://issues.apache.org/jira/browse/SQOOP-1920?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14252941#comment-14252941
]
Veena Basavaraj commented on SQOOP-1920:
----------------------------------------
[~hshreedharan] suggested using
https://hadoop.apache.org/docs/current/api/org/apache/hadoop/metrics2/package-summary.html
for MR engine.
> Ability to expose more metrics of the joblifecyle ( related to the connectors)
> ------------------------------------------------------------------------------
>
> Key: SQOOP-1920
> URL: https://issues.apache.org/jira/browse/SQOOP-1920
> Project: Sqoop
> Issue Type: Sub-task
> Reporter: Veena Basavaraj
> Fix For: 1.99.6
>
>
> This also ties in with how we will support failure handling in the Job
> execution flow SQOOP-1856, we have one counter / metric today that says
> number of rows read, it would be nice ( in fact complete) to track how many
> rows were written, the time taken to do it and few more things that will give
> visibility into what happened in side the execution engine, are hadoop
> counters the way to go?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
