[
https://issues.apache.org/jira/browse/HBASE-5828?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13269543#comment-13269543
]
xufeng commented on HBASE-5828:
-------------------------------
I just review the code and have a speculation about this issue:
If the region of META table assigned on regionserver_A.
The region of user table assigned on regionserver_B.
At first,the pipeline of regionserver_A and regionserver_B have the same
datanodes.
The testcase testLogRollOnDatanodeDeath will kill the all datanodes in pipeline
which belongs to the hlog writer of regionserver_B.
When the testcase testLogRollOnPipelineRestart to deleteTale,it will update the
META data.
Because the datanodes in pipeline of regionserver_A have been stoped,So it will
fail.
Test will wait in HBaseAdmin#deletaTable() until retries exhausted.
After long time this test will be killed.
> TestLogRolling fails in 0.90 branch killing the test suite up on jenkins
> ------------------------------------------------------------------------
>
> Key: HBASE-5828
> URL: https://issues.apache.org/jira/browse/HBASE-5828
> Project: HBase
> Issue Type: Bug
> Reporter: stack
>
> See recent 0.90 builds up on jenkins:
> https://builds.apache.org/view/G-L/view/HBase/job/hbase-0.90/471/console
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
