On Aug 16, 2013, at 5:30 AM, Chris Filo Gorgolewski <krzysztof.gorgolewski@xxxxxxxxx> wrote:
When jobs hit memory limits, they are put into the HOLD state with an appropriate message and hold code. If you want them to be re-run, you could use PeriodicHold. If you want the job to be automatically edited by the system - you may want to look at the JobRouter, which can do periodic job edits.
That said, I think it's much better to hold the job and let the user examine it manually. If they aren't informed of the issue, they may never be aware of it (and waste thousands of CPU hours - I've seen it!).