[
https://issues.apache.org/jira/browse/HBASE-14362?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14934427#comment-14934427
]
Heng Chen commented on HBASE-14362:
-----------------------------------
If i set param
{code}
conf.setInt("hbase.procedure.store.wal.max.roll.retries", 1);
conf.setInt("hbase.procedure.store.wal.sync.failure.roll.max", 1);
{code}
This issue will be reproduced locally.
I think the reason is, as original logic
{code}
store.insert(new TestProcedure(i, -1), null);
waitForNumReplicas(3);
{code}
after we restart dn, we insert immediately, when dn is not started fully, the
testcase will failed.
So we use {{waitForNumReplicas(3)}} before insert, in
{{waitForNumReplicas(3)}} it will wait dn start fully.
after do that, the testcase will not failed again locally.
> org.apache.hadoop.hbase.master.procedure.TestWALProcedureStoreOnHDFS is super
> duper flaky
> -----------------------------------------------------------------------------------------
>
> Key: HBASE-14362
> URL: https://issues.apache.org/jira/browse/HBASE-14362
> Project: HBase
> Issue Type: Bug
> Components: test
> Affects Versions: 2.0.0
> Reporter: Dima Spivak
> Priority: Critical
> Attachments: HBASE-14362.patch
>
>
> [As seen in
> Jenkins|https://builds.apache.org/job/HBase-TRUNK/lastCompletedBuild/testReport/org.apache.hadoop.hbase.master.procedure/TestWALProcedureStoreOnHDFS/history/],
> this test has been super flaky and we should probably address it.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
