[
https://issues.apache.org/jira/browse/HBASE-15264?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15153951#comment-15153951
]
Duo Zhang commented on HBASE-15264:
-----------------------------------
Oh, I forgot an important thing, we can not close WAL writer asynchronously and
also we can not use recoverFileLease.
The reason is that, we use the lease on WAL file to do fencing, which means
that, if we failed to close the WAL file current being written , it is likely
that we are considered dead on master.
So the close must be synchronous, and also, we should close the file normally
instead of calling recoverFileLease...
Let me retry the recoverAndClose part...
Thanks.
> Implement a fan out HDFS OutputStream
> -------------------------------------
>
> Key: HBASE-15264
> URL: https://issues.apache.org/jira/browse/HBASE-15264
> Project: HBase
> Issue Type: Sub-task
> Components: util, wal
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Fix For: 2.0.0
>
> Attachments: HBASE-15264-v1.patch, HBASE-15264-v2.patch,
> HBASE-15264-v3.patch, HBASE-15264-v4.patch, HBASE-15264.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
