hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Harsh J (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-3918) EditLogTailer shouldn't log WARN when other node is in standby mode
Date Tue, 19 Feb 2013 18:03:13 GMT

    [ https://issues.apache.org/jira/browse/HDFS-3918?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13581488#comment-13581488
] 

Harsh J commented on HDFS-3918:
-------------------------------

Change looks good, but just so that the string change not cause an issue later (since we are
apparently splitting it), can we have some form of regression-catch to ensure [0] is always
what we want it to be?

Not very necessary, but would be good to have. If its tough to do, we can punt. +1 with this
addressed (in reply or in patch).
                
> EditLogTailer shouldn't log WARN when other node is in standby mode
> -------------------------------------------------------------------
>
>                 Key: HDFS-3918
>                 URL: https://issues.apache.org/jira/browse/HDFS-3918
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>         Attachments: hdfs-3918.txt
>
>
> If both nodes are in standby mode, each will be trying to roll the others' logs, which
results in errors like:
> Unable to trigger a roll of the active NN 
> org.apache.hadoop.ipc.StandbyException: Operation category JOURNAL is not supported in
state standby
> We should catch this specific exception and not log it at WARN level, since it's expected
behavior.

--
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

Mime
View raw message