Counters and other useful stats should be logged into Job History log
---------------------------------------------------------------------
Key: HADOOP-2581
URL: https://issues.apache.org/jira/browse/HADOOP-2581
Project: Hadoop
Issue Type: Improvement
Components: mapred
Reporter: Runping Qi
The following stats are useful and available to JT but not logged job history
log:
1. The counters of each job
2. The counters of each mapper/reducer attempt
3. The info about the input splits (filename, split size, on which nodes)
3. The input split for each mapper attempt
Those data is useful and important for mining to find out performance related
problems.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
