[cachebox] Cache.reapAll() Causing Thread Overloads

Anyone run into this? Our ColdFusion server is throwing tons of errors due to the ReapAll() in the OnRequestStart, which is recommended when used as standalone. It seems to run out of threads. “coldfusion.util.GenericThreadPool$QueueFullException: Cannot create a new thread because the task queue has reached it maximum limit 5000”. The only other thing we used cfthread for is a daily nighttime task. I commented out the line and the errors stopped but as soon as I tried to re-enabled it they started coming in again.

Mary Jo Sminkey

Does a full JVM thread dump show if more than one reap thread is running? It sounds like the reaping is running over and over and never completing.



ColdBox/CommandBox Developer Advocate
Ortus Solutions, Corp

E-mail: brad@coldbox.org
ColdBox Platform: http://www.coldbox.org
Blog: http://www.codersrevolution.com

I’d assume so as well, but I’d have to ask one of our network guys to check the next time it happens. The question still would be, why it’s doing that and if there’s anything that would help. I am guessing maybe bumping up the Reap Threshold? This is about the 3rd time in a couple months that this has happened and where the server had to be restarted to get past the errors.

Mary Jo

Just to clarify, we don’t normally see these. When they do occur, the server starts throwing them on every request and the server has to be restarted.