hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stack <st...@duboce.net>
Subject Re: precommit zombies
Date Wed, 09 Jan 2013 15:58:03 GMT
On Wed, Jan 9, 2013 at 7:17 AM, Nicolas Liochon <nkeywal@gmail.com> wrote:

> FYI, We sometimes have remains of other product builds.

You think these mess up our tests at all N?  Or are these just the zombies
sometimes reported in our test output on jenkins?

> Lastly, HDFS,
> likely HDFS-4261 <https://issues.apache.org/jira/browse/HDFS-4261>.
>    java.lang.Thread.State: TIMED_WAITING (sleeping)
>         at java.lang.Thread.sleep(Native Method)
>         at
> org.apache.hadoop.hdfs.server.balancer.Balancer.run(Balancer.java:1457)
>         at
> org.apache.hadoop.hdfs.server.balancer.TestBalancerWithNodeGroup.runBalancer(TestBalancerWithNodeGroup.java:170)
>         at
> org.apache.hadoop.hdfs.server.balancer.TestBalancerWithNodeGroup.testBalancerWithRackLocality(TestBalancerWithNodeGroup.java:220)
> The Jenkins process should clean this. I don't know if we can do it
> ourselves automatically before starting our tests?:
> - may be it's not the same user for the different builds
> - we could kill legitimates Java process.

Want to add a listing of running java processes before test starts?  You
have admin privs on jenkins Nicolas?  I can add.

Selective kill seems like good idea if jenkins is not doing it (long as we
don't kill jenkins itself --- that'd make us popular... smile)


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message