Hi, Jenkins slaves on pharos-contribution often go off-line because their disk is full. There are a lot of job executed there and the jenkins workspace on each slave growth quickly. To avoid that, I set up a job to delete the whole workspace on slaves once a week: https://ci.inria.fr/pharo-contribution/job/purge-jenkins-slaves-workspace. It ghouls solve this issue but has a drawback: sometimes, you will not have access to the workspace of the latest build of a job. In fact, it is not a problem if you archive artefacts at the end of a build.
Christophe
