[
https://issues.apache.org/jira/browse/HBASE-10000?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13846019#comment-13846019
]
Ted Yu commented on HBASE-10000:
--------------------------------
Stack:
I understand the importance of this part of the code.
When HBASE-7878 was checked in, isFileClosed() was not available in hadoop
release. That limited our capability of verifying whether lease has been
recovered.
The patch kept existing semantics of lease recovery - as you can see from the
new unit test.
One addition w.r.t. yesterday's test was that there were about 20 HLog files
being recovered around the time I killed that region server. These were due to
concurrent system tests.
Let me increase the scale of testing after addressing one more round of code
review.
> 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
> Fix For: 0.98.1
>
> Attachments: 10000-0.96-v5.txt, 10000-0.96-v6.txt,
> 10000-recover-ts-with-pb-2.txt, 10000-recover-ts-with-pb-3.txt,
> 10000-recover-ts-with-pb-4.txt, 10000-recover-ts-with-pb-5.txt,
> 10000-recover-ts-with-pb-6.txt, 10000-recover-ts-with-pb-7.txt,
> 10000-recover-ts-with-pb-8.txt, 10000-recover-ts-with-pb-8.txt, 10000-v4.txt,
> 10000-v5.txt, 10000-v6.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
(v6.1.4#6159)
