From time to time I need to restart some job on same machine even if it wasn't initially designed for any specific system, just a job for random machine. Sometimes unexpected issue occurs on a hardware and I have to troubleshoot it there. Simplest use case is kernel panic. If this occurs I want to double-check if there is easy reproducer on that specific machine and how it is affected by changing some attributes of job.
Bulk reassignment of issues as Bill has moved to another team.
This would be relevant to the planned automatic rescheduling feature as well.