After rebooting our servers problem will be immediately reproduced on another node. There are some JMS Connection leaks, but those unnecessary connections will be closed after 30 minutes (channel-idle-timeout-millis=30 minutes). 1. There are not any high amount of JMS connections. 2. There is no high amount of JMS Messages.
So have anyone faced a similar issue with GraniteDS causing high CPU usage? May be you may have some advices or places to look for?
Thank you in advance.