hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Himanshu Vashishtha (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-7723) Remove NN URI from ZK splitlogs.
Date Fri, 01 Feb 2013 00:49:13 GMT

    [ https://issues.apache.org/jira/browse/HBASE-7723?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13568306#comment-13568306

Himanshu Vashishtha commented on HBASE-7723:

Yea, that's the reason why the above tests failed. I am thinking to do:
a) In SplitLogManager, while inserting the tasks, chomp off the hbase.rootDir prefix from
the filename. This will make the path relative from the .logs dir (or whatever the user has
used). Use this for the znodes.

b) For the SplitLogWorkers: Either make the condition to clear out the existing znodes a necessary
condition. This way, we don't have to worry about old znodes with NN URI. Or, re-create splitlog
znode when a master becomes active (recursively delete the znode). This might interrupt with
the regionservers doing currently the DLS as the task znode is no longer there. But then,
the MasterFileSystem.splitLog() will make sure to re-enque the tasks.

> Remove NN URI from ZK splitlogs.
> --------------------------------
>                 Key: HBASE-7723
>                 URL: https://issues.apache.org/jira/browse/HBASE-7723
>             Project: HBase
>          Issue Type: Bug
>          Components: hadoop2, master
>    Affects Versions: 0.92.0
>            Reporter: Kevin Odell
>            Assignee: Himanshu Vashishtha
>         Attachments: HBASE-7723-94.patch
> When moving to HDFS HA or removing HA we end up changing the NN namespace.  This can
cause the HMaster not to start up fully due to trying to split phantom HLogs pointing to the
wrong FS - java.lang.IllegalArgumentException: Wrong FS: error messages.  The HLogs in question
might not even be on HDFS anymore.  You have to go in a manually clear out the ZK splitlogs
directory to get HBase to properly boot up.

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

View raw message