deadlineslave.exe enormous memory consumption.

We have a situation thats been building on the farm for awhile. Our deadlineslave.exe files are hoarding a ton of memory and I can’t find a reason why or a solution to clear them out except restart. But nothing automatic and we have quite a lot of nodes to work thru.

Is this normal? Anything to do about it? It doesn’t seem to be specific scenes that cause this problem. Anything we can test? It can keep memory in there when no jobs are running on the machine.

I think this is where all our stalled slaves error are coming from and also Error in StartJob: Timed out waiting for the next progress update.

We are on Deadline 6.0

Thanks for any incite!

-John

Hey John,

We believe this issue has been resolved in Deadline 6.1. In Deadline 6.0, the application plugins and event plugins would leak a bit after each job, and we added a new Cleanup callback to all the plugins that would prevent this. So if you can upgrade to 6.1, that should solve your problem.

Cheers,
Ryan