[
https://issues.apache.org/jira/browse/FALCON-133?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13777402#comment-13777402
]
Srikanth Sundarrajan commented on FALCON-133:
---------------------------------------------
Hi Jean, Agree that this simplifies logging and any superfluous string concats.
But some of changed code actually runs on hadoop. For example FeedEvictor code
actually runs on the hadoop cluster as a map-reduce job. The Hadoop
installation already includes slf4j (1.4.3 - Hadoop 1.* & 1.6.1 - Hadoop 2.*).
Would this create a runtime conflict (haven't checked myself to see if 1.7.5 is
fully backward compatible with 1.4.3 & later) ? Can you please confirm if this
doesn't break anything particularly retention, replication, recordsize and
post-processing actions? If there are no conflicts, this would be a good change.
Thanks
> Upgrade to slf4j 1.7.5 and use SLF4J logger
> -------------------------------------------
>
> Key: FALCON-133
> URL: https://issues.apache.org/jira/browse/FALCON-133
> Project: Falcon
> Issue Type: Improvement
> Components: general
> Reporter: Jean-Baptiste Onofré
> Fix For: 0.4
>
> Attachments: FALCON-133.patch
>
>
> SLF4J 1.7.5 provides interesting performance improvements.
> More over, it's not more require to test the log level, we can use directly
> {} in the message formatter.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
