This one is successful again, but just because it ran on different Jenkins node (ubuntu1) which had enough disk space, compared to Jenkins node where it regularly fails (solaris1) which doesn't have that much disk space.
MAHOUT-1275 will help with this but there are more things we can do about how much disk space each build takes, and reduce disk space footprint of all of our build jobs. E.g. for this build job it seems that archiving build artifacts by Jenkins is turned on - if I'm not mistaken we do not promote Jenkins builds into releases, or use anything else that could make use of archived build artifacts, so IMO this archiving can be turned off. Also, we can in existing build jobs add pre build step (or modify pom to have a new "ci" profile and in it), to execute a Maven build which will cleanup all the previous builds of Mahout from Jenkins node local maven repository with the help of build helper maven plugin and its remove-project-artifact goal). This can be tested on solaris1 Jenkins node, by configuring nightly build job to temporarily execute only that node only. What do you guys think of these build improvement ideas? I guess I should create separate JIRA tickets for these two if they seem reasonable. Who has necessary rights to make these changes, or how does one get them granted? Kind regards, Stevo Slavić. On Wed, Jul 10, 2013 at 2:03 AM, Apache Jenkins Server < jenk...@builds.apache.org> wrote: > See < > https://builds.apache.org/job/mahout-nightly/org.apache.mahout$mahout-integration/1288/ > > > >
