[
https://issues.apache.org/jira/browse/HBASE-3130?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13105743#comment-13105743
]
Jean-Daniel Cryans commented on HBASE-3130:
-------------------------------------------
Some comments on the contribution itself since you're a newcomer:
- The logging should be done like we do it elsewhere in the code, by using the
Log class provided by Commons Logging and then using the LogFactory.
ReplicationSource is an example.
- When logging, try to include relevant information because with all different
threads logging it can be hard to follow. For example, this could tell which
peer we're talking about:
bq. Log.info("Refreshing ZookeeperWatcher");
- Lines should be max 80 characters long (the one where you're catching
SessionExpiredException for example).
More info in this chapter of the book:
http://hbase.apache.org/book/developer.html
Now on the content, I still think that the management of the connection should
be done inside ReplicationZookeeper. In its current form the patch exposes a
feature envy between ReplicationSource and ReplicationPeer, might just be
better to contain the functionality in between.
Thanks for working on this Chris!
> [replication] ReplicationSource can't recover from session expired on remote
> clusters
> -------------------------------------------------------------------------------------
>
> Key: HBASE-3130
> URL: https://issues.apache.org/jira/browse/HBASE-3130
> Project: HBase
> Issue Type: Bug
> Components: replication
> Affects Versions: 0.92.0
> Reporter: Jean-Daniel Cryans
> Assignee: Chris Trezzo
> Fix For: 0.92.0
>
> Attachments: 3130.txt
>
>
> Currently ReplicationSource cannot recover when its zookeeper connection to
> its remote cluster expires. HLogs are still being tracked, but a cluster
> restart is required to continue replication (or a rolling restart).
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
