A stupid simple script to check for problems with job data on disk.
This was written to combat problems we've been having with builds disappearing after the build completed as well as some other problems we've had over the years (cruft can really build up in your jobs!).
Point it at a directory in $JENKINS_HOME/jobs and it will scan that
job's build history:
$ ruby jobber.rb "$JENKINS_HOME/jobs/MyJob"It will then print out any problems it finds and solutions it proposes.
It is reasonable to run it on all jobs:
$ ruby jobber.rb "$JENKINS_HOME/jobs/"*WARNING I tried my best not to break anything. You do have good backups, right?
You can let jobber.rb try to fix those problems itself:
$ ruby jobber.rb "$JENKINS_HOME/jobs/MyJob" --solveIt'll then run its proposed solutions.
Note: You need to either restart Jenkins, "reload from disk", or
better yet, run jobber.rb --solve while Jenkins is down.