I am running Talend Open Studio for ESB 6.3 with the route deployed in Runtime Environment (karaf) that is triggered via cTimer and run cTalendJob every 30 seconds. Talend Job then invokes another Talend Jobs using tRunJob.
Heap space is constantly growing, and when it reaches maximum allocated 8GB everything slows down, java is consuming 100% CPU, and I have to restart the service and start it again (wrapper is in use as windows service). It takes about 1 week to consume all 8GB.
I did memory analyze using Eclipse Memory Manager. Dump is taken just 2 days after restart, suspected problem is job started from route:
The classloader/component "local_project.MyJob" occupies 231 335 472 (58,54 %) bytes. The memory is accumulated in one instance of "java.util.HashMap$Node[]" loaded by "<system class loader>".
Anyone with idea why memory is not released after cTalendJob in route is completed?
Thank you in advance
origami
