GNU Queue load-balancing system:How does Queue compare with the commercial LSF product?:Does Queue allow (optional) automated resubmission of a job that dies due to a downed machine like LSF does?
|
Partially. Queued only deletes a job on successful completion (i.e., it died; queued didn't go down), so if the machine goes down it should be restarted on the same machine when queued restarts. What is needed is a
feature to turn this off.
What won't happen (I thought about this sense someone asked a question about this earlier) is it (probably) won't start up on another machine, because other queued will mark the jobs as "ANOTHER_HOST" when
they sense it has been locked by another machine.
|
Does Queue have commands like LSF to administrator commands to suspend/manipulate/monitor any/all jobs or queues within the system?