www-builds mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Uwe Schindler" <...@thetaphi.de>
Subject RE: Lucene Jenkins slave out of disk
Date Thu, 21 Apr 2011 07:42:54 GMT

I was not aware that you took action, too. Because there is the message
about Jenkins shutdown, I was assuming that you are fixing the master.

About the Lucene slave, the problem we have is: Our tests need lots of disk
space for large indexes. Before the build, the workspace is complete cleared
(at least for the nightly builds) and for the half-hourly builds it is
svn-upped and then cleaned by ANT (ant clean). But at the end of the build
the build script does not call "ant clean" again, because otherwise we would
lost the test results. I am working on adding an additional task to our ant
build scripts that will only clean up the test data that can be called after
the build finished. This may clean up unneeded disk space after build so
later running builds can reuse the space.

I will talk to the other committers who know better where the temporary test
files lie.


Uwe Schindler
H.-H.-Meier-Allee 63, D-28213 Bremen
eMail: uwe@thetaphi.de

> -----Original Message-----
> From: Niklas Gustavsson [mailto:niklas@protocol7.com]
> Sent: Thursday, April 21, 2011 8:57 AM
> To: builds@apache.org
> Cc: dev@lucene.apache.org
> Subject: Re: Lucene Jenkins slave out of disk
> On Thu, Apr 21, 2011 at 8:42 AM, Uwe Schindler <uwe@thetaphi.de> wrote:
> > [root@lucene /home/hudson/hudson-slave/workspace]# df -h
> Filesystem
> > Size    Used   Avail Capacity Mounted on
> > zroot/jails/lucene.zones.apache.org     69G     53G     17G    76%
> > /
> >
> > I think the master is this time out of disk space.
> The disk space was cleared (or at least that's what Jenkins thinks).
> The slave is now running again.
> /niklas

View raw message