I accidentally had two run-jobs processes going at the same time, and a job ended up stuck in in_process forever. I have no idea why, but that seems extremely poor - jobs are supposed to be extremely isolated such that any number of runners could theoretically be going. This isn't generally useful to do in practice since the most painful jobs have to read/write to a central filesystem. But it should still be possible to have multiple runners without bugs.
I accidentally had two
run-jobsprocesses going at the same time, and a job ended up stuck inin_processforever. I have no idea why, but that seems extremely poor - jobs are supposed to be extremely isolated such that any number of runners could theoretically be going. This isn't generally useful to do in practice since the most painful jobs have to read/write to a central filesystem. But it should still be possible to have multiple runners without bugs.