[jira] [Commented] (HDFS-5936) TestCacheDirectives#testCacheManagerRestart is flaky

["Binglin Chang (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HDFS-5936?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13911424#comment-13911424 ] 

Binglin Chang commented on HDFS-5936:
-------------------------------------

I got the same failure in HDFS-6002, I suspect the reason for the failure is dir for seconderynamenode is not deleted in previous tests, so when this test run into CheckpointSignature.validateStorageInfo, the namespaceId and blockpoolId cannot match. 
Normally a new created MInicluster will deleted the old namedir and format a new one, but it does not cleanup namesecondery dir. I cannot verify this caused the failure, but it looks like a bug anyway.

> TestCacheDirectives#testCacheManagerRestart is flaky
> ----------------------------------------------------
>
>                 Key: HDFS-5936
>                 URL: https://issues.apache.org/jira/browse/HDFS-5936
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: caching
>    Affects Versions: 2.3.0
>            Reporter: Andrew Wang
>
> Seen this happen on a few Jenkins precommit runs:
> {code}
> java.io.IOException: Inconsistent checkpoint fields.
> LV = -52 namespaceID = 1980502516 cTime = 0 ; clusterId = testClusterID ; blockpoolId = BP-942435048-67.195.138.22-1392086493767.
> Expecting respectively: -52; 2; 0; testClusterID; BP-410204915-67.195.138.22-1392086484908.
> 	at org.apache.hadoop.hdfs.server.namenode.CheckpointSignature.validateStorageInfo(CheckpointSignature.java:133)
> 	at org.apache.hadoop.hdfs.server.namenode.SecondaryNameNode.doCheckpoint(SecondaryNameNode.java:526)
> 	at org.apache.hadoop.hdfs.server.namenode.TestCacheDirectives.testCacheManagerRestart(TestCacheDirectives.java:582)
> {code}



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)