accumulo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dylan Hutchison <dhutc...@cs.washington.edu>
Subject Re: Orphaned MAC processes
Date Wed, 31 Aug 2016 16:13:36 GMT
I have noticed it many times in the past.  I think they popped up when I
launched a test that crashed the tserver with an exception.  When I was
doing development this happened often enough that I made a script to do
grep and kill them.  Under normal circumstances they die as normal.

On Wed, Aug 31, 2016 at 8:41 AM, Josh Elser <josh.elser@gmail.com> wrote:

> Tried running the ITs overnight for 1.8.0rc3
>
> {noformat}
> Running org.apache.accumulo.test.replication.UnorderedWorkAssignerRe
> plicationIT
> OpenJDK 64-Bit Server VM warning: INFO: os::commit_memory(0x00000000f1900000,
> 157810688, 0) failed; error='Cannot allocate memory' (errno=12)
> #
> # There is insufficient memory for the Java Runtime Environment to
> continue.
> # Native memory allocation (malloc) failed to allocate 157810688 bytes for
> committing reserved memory.
> # An error report file with more information is saved as:
> # /home/elserj/accumulo-180-rc3/accumulo-1.8.0-src/test/hs_err_pid5342.log
> Running org.apache.accumulo.test.replication.KerberosReplicationIT
> Running org.apache.accumulo.test.replication.ReplicationRandomWalkIT
> Running org.apache.accumulo.test.replication.SequentialWorkAssignerIT
> Running org.apache.accumulo.test.replication.GarbageCollectorCommuni
> catesWithTServersIT
> Running org.apache.accumulo.test.replication.WorkMakerIT
> zsh: killed     mvn verify -fn
> {noformat}
>
> What I'm seeing now is that I have 28 orphaned Java processes, all of them
> appear to be from MAC on these specific test classes.
>
> Has anyone else noticed that we're leaking these processes? Are we not
> launching them correctly such that when the PPID dies, they continue to
> live?
>

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