Duo Zhang commented on HBASE-19964:

OK I think this is something happened before. Output the memstore size after 
the compaction check, it is not zero. This means that we fail to flush the 
table. Change to use HBaseTestingUtility.flush can fix the problem.

But anyway, I'm still a little worried about the Admin.flush, we may fail to 
flush without any notification to users.

> TestWriteHeavyIncrementObserver fails
> -------------------------------------
>                 Key: HBASE-19964
>                 URL: https://issues.apache.org/jira/browse/HBASE-19964
>             Project: HBase
>          Issue Type: Bug
>          Components: Coprocessors
>            Reporter: Duo Zhang
>            Priority: Major
>             Fix For: 2.0.0-beta-2
> https://builds.apache.org/job/HBASE-Flaky-Tests/25704/artifact/hbase-examples/target/surefire-reports/org.apache.hadoop.hbase.coprocessor.example.TestWriteHeavyIncrementObserver-output.txt/*view*/
> I could reproduce it locally. Let me dig more.

This message was sent by Atlassian JIRA

Reply via email to