iwasakims opened a new pull request #2507:
URL: https://github.com/apache/hadoop/pull/2507


   https://issues.apache.org/jira/browse/HDFS-15702
   
   dfs.blockreport.intervalMsec is set to 1000 in the TestDecommission. While 
the testAllocAndIBRWhileDecommission tries to stop IBRs by 
DataNodeTestUtils#pauseIBR, IBR still can be sent in the code path of FBR 
(BPServiceActor#blockReport). Avoiding FBR (except for the 1st BR) by setting 
dfs.blockreport.intervalMsec to enough long value should fix the issue. The 
testAllocAndIBRWhileDecommission does not depend on FBR.
   
   The test failure can not be reproduced in 100 runs after applying the patch 
on my local.
   
   ```
   for i in `seq 100` ; do echo $i && mvn test -DignoreTestFailure=false 
-Dtest=TestDecommission#testAllocAndIBRWhileDecommission || break ; done
   ```
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to