Description of problem: The https://build.gluster.org/job/netbsd-periodic/ job generates a .tgz on failure (when a new core was detected), but it also contains all the other cores that are under build/install/cores/ . This makes the archive larger than needed, and more difficult to find the correct core. It would help tremendously if: - old cores can be deleted before running the tests - a backtrace is logged in the Jenkins console output (and maybe written to a file in the archive), just like for centos6-regression
The job is not running anymore, and not planning to support complete regression on NetBSD for now.