If you guys talk about https://builds.apache.org/job/Jackrabbit-trunk/2116/console You won't fix that with MAVEN_OPTS, when running surefire a new jvm is forked (which don't use MAVEN_OPTS). IMHO/Maybe there are some zombies process running in the solaris instances which eat the memory. I cannot log to the machine now but that's the possible trouble.
2013/5/7 "Andreas Lehmkühler" <[email protected]>: > Hi, > > Jukka Zitting <[email protected]> hat am 7. Mai 2013 um 11:32 > geschrieben: >> Hi, >> >> [cc += [email protected]] >> >> On Tue, May 7, 2013 at 11:26 AM, Apache Jenkins Server >> <[email protected]> wrote: >> > Check console output at >> > https://builds.apache.org/job/Jackrabbit-trunk/2116/ >> > to view the results. >> >> The problem is: >> >> [INFO] Error while executing forked tests.; nested exception is >> org.apache.maven.surefire.booter.shade.org.codehaus.plexus.util.cli >> .CommandLineException: Error setting up environmental variables >> error=12, Not enough space >> >> I'm not sure if this is caused by the solaris1 slave running out of >> memory or the Jackrabbit-trunk build just not having enough memory to >> work with. > We have the same problem with our PDFBox build. Thanks for the hint, at first > I thought the error is disc space related .... > >> For now I updated the build to use MAVEN_OPTS="-Xmx1g >> -XX:MaxPermSize=300m" to cover for the latter case. Let's see what >> happens. > It did the thr trick for us, at least for our last build. > > Thanks again for the input! > >> BR, >> >> Jukka Zitting > > > BR > Andreas Lehmkühler -- Olivier Lamy Ecetera: http://ecetera.com.au http://twitter.com/olamy | http://linkedin.com/in/olamy
