[jira] [Commented] (HBASE-6904) In the HBase shell, an error is thrown that states replication-related znodes already exist

["Gregory Chanan (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HBASE-6904?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13487353#comment-13487353 ] 

Gregory Chanan commented on HBASE-6904:
---------------------------------------

The issue is in RecoverableZooKeeper.createNonSequential handling of NodeExistsException:
{code}
            LOG.error("Node " + path + " already exists and this is not a " +
                "retry");
            throw e;
{code}

The hbase shell reprints log messages at error+ level.  This error log is unnecessary, because the code already throws an exception.  The code that calls it can just decide if it wants to reprint the error message or not by catching/rethrowing the exception.  I looked through the calls, which all originate in ZKUtil, and I think it's fine to just bump this down to "info" level.  All the calls in ZKUtil either rethrow the exception or have semantics where getting a NodeExists exception is not a problem.
                
> In the HBase shell, an error is thrown that states replication-related znodes already exist
> -------------------------------------------------------------------------------------------
>
>                 Key: HBASE-6904
>                 URL: https://issues.apache.org/jira/browse/HBASE-6904
>             Project: HBase
>          Issue Type: Bug
>          Components: Replication, Zookeeper
>    Affects Versions: 0.92.0, 0.92.1
>            Reporter: Aleksandr Shulman
>            Priority: Minor
>
> On a replication-enabled cluster, querying the list_peers produces the error lines shown below. It doesn't appear that anything is broken in terms of functionality.
> Stack trace:
> hbase(main):001:0> list_peers
> 12/09/29 14:41:03 ERROR zookeeper.RecoverableZooKeeper: Node /hbase/replication/peers already exists and this is not a retry
> 12/09/29 14:41:03 ERROR zookeeper.RecoverableZooKeeper: Node /hbase/replication/rs already exists and this is not a retry
> PEER ID CLUSTER KEY
> 0 row(s) in 0.4650 seconds

--
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