Description of problem: Due to BZ#967479, host and guest recipes that cannot be satisfied on the same lab controller may stay queued forever. We need a way of aborting them in a timely fashion. Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. submit a host/guest recipeset, where the guest distro tree is not available in any lab where the host's distro tree is. 2. 3. Actual results: It will stay queued forever, or until the distro becomes available in the relevant lab controller. Expected results: It should wait for what is a reasonable amount of time (this should be configurable) in the queued state (due to the distro not being available) before it is aborted in the abort_dead_recipes() loop. Additional info:
Fixing bug 967479 seems to generally avoid this problem in practice. Marking this as "WONTFIX" unless/until someone indicates that it is causing a real issue.