[
https://issues.apache.org/jira/browse/HBASE-20839?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16531066#comment-16531066
]
Duo Zhang commented on HBASE-20839:
-----------------------------------
Manually tested with the following command
{noformat}
mvn clean test -Dtest=TestSequenceIdMonotonicallyIncreasing
-Dhadoop-two.version=2.8.4
{noformat}
Without the patch, the test will hang. With the patch in place, the test can
finish normally. And in log we can see this
{noformat}
2018-07-03 16:57:55,104 INFO [RS:0;zhangduo-ubuntu:45551] wal.WALFactory(148):
Instantiating WALProvider of type class
org.apache.hadoop.hbase.wal.FSHLogProvider
2018-07-03 16:57:55,104 INFO [RS:1;zhangduo-ubuntu:41447] wal.WALFactory(148):
Instantiating WALProvider of type class
org.apache.hadoop.hbase.wal.FSHLogProvider
{noformat}
Which means we successfully fall back to use FSHLog.
> Fallback to FSHLog if we can not instantiated AsyncFSWAL when user does not
> specify AsyncFSWAL explicitly
> ---------------------------------------------------------------------------------------------------------
>
> Key: HBASE-20839
> URL: https://issues.apache.org/jira/browse/HBASE-20839
> Project: HBase
> Issue Type: Sub-task
> Components: wal
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Priority: Blocker
> Fix For: 3.0.0, 2.1.0, 2.0.2, 2.2.0
>
> Attachments: HBASE-20839.patch
>
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
