Hello, We have a rather large workflow (120 steps) that works on one of our Galaxy instances, but crashes Galaxy on another instance. All the individual tools work on both instances. The error in paster.log is: TimeoutError: QueuePool limit of size 5 overflow 10 reached, connection timed out, timeout 30 This is an older linux machine running CentOS, and there is no queuing system. Is the queuepool limit a Galaxy variable? Or is this a limitation of how we have the machine configured? I've tried searching the Galaxy wiki, but no luck yet. I will keep searching, too. Thanks! Todd -- *************************************** Todd Oakley, Professor Ecology Evolution and Marine Biology University of California, Santa Barbara Santa Barbara, CA 93106 USA ***************************************