[
https://issues.apache.org/jira/browse/HBASE-10000?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13830154#comment-13830154
]
Ted Yu commented on HBASE-10000:
--------------------------------
I am currently in China.
Will test on cluster once I come up with next rev of the patch - patch should
have used a threadpool in submitting recoverLease() calls.
Another improvement I want to try, maybe in another JIRA, is to let split log
worker to pick up oldest WAL file(s) first where chance of getting an already
closed file is higher. Himanshu mentioned this idea in HBASE-9873.
> Initiate lease recovery for outstanding WAL files at the very beginning of
> recovery
> -----------------------------------------------------------------------------------
>
> Key: HBASE-10000
> URL: https://issues.apache.org/jira/browse/HBASE-10000
> Project: HBase
> Issue Type: Improvement
> Reporter: Ted Yu
> Assignee: Ted Yu
> Attachments: 10000-v1.txt
>
>
> At the beginning of recovery, master can send lease recovery requests
> concurrently for outstanding WAL files using a thread pool.
> Each split worker would first check whether the WAL file it processes is
> closed.
> Thanks to Nicolas Liochon and Jeffery discussion with whom gave rise to this
> idea.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
