You are viewing a plain text version of this content. The canonical link for it is here.

Posted to builds@apache.org by Jesse Glick <je...@oracle.com> on 2010/07/06 19:48:21 UTC

Possibly stalled jobs

Several jobs - Zookeeper-Patch-h1.grid.sp2.yahoo.net, Pig-Patch-h8.grid.sp2.yahoo.net, Tuscany-2x - are extremely slow (running for several hours now), and history shows 
that they were often just aborted. I think these should be given a timeout. Tuscany-2x does actually have a timeout of 400 minutes, but that seems excessively generous 
for a job tying up one of two executors on one of the only Solaris slaves. (BTW why is hudson-solaris labeled "Solaris" rather than "Solaris 10" like 
lucene.zones.apache.org, when it in fact appears to be running Solaris 10?)

Re: Possibly stalled jobs

Posted by Justin Mason <jm...@jmason.org>.

hey -- the yahoo.net executors are dedicated to Hadoop-related
projects, so shouldn't affect the "general-purpose" jobs.  Agreed
though that 400 minutes seems excessive.  Are any Tuscany-related
folks on the list to discuss?

--j.

On Tue, Jul 6, 2010 at 18:48, Jesse Glick <je...@oracle.com> wrote:
> Several jobs - Zookeeper-Patch-h1.grid.sp2.yahoo.net,
> Pig-Patch-h8.grid.sp2.yahoo.net, Tuscany-2x - are extremely slow (running
> for several hours now), and history shows that they were often just aborted.
> I think these should be given a timeout. Tuscany-2x does actually have a
> timeout of 400 minutes, but that seems excessively generous for a job tying
> up one of two executors on one of the only Solaris slaves. (BTW why is
> hudson-solaris labeled "Solaris" rather than "Solaris 10" like
> lucene.zones.apache.org, when it in fact appears to be running Solaris 10?)
>
>