Hide Forgot
Description of problem: Candlepin 2.0.11 is deployed in the IT-PnT development environment on candlepin06.web.dev.int.devlab.redhat.com. A few requests is enough to render the tomcat unresponsive. We suspect a memory leak Version-Release number of selected component (if applicable): 2.0.11 How reproducible: Becomes unresponsive approximately an hour after a restart. Restart seems to clear it up temporarily Steps to Reproduce: 1. Start tomcat 2. Make status requests and pool refresh requests 3. Observe lengthening response times Actual results: Requests eventually time out due to an unresponsive server Expected results: Requests should be handled normally Additional info: Heap dump from unresponsive state attached
Created attachment 1152326 [details] Heap dump from candlepin06 while tomcat was in unresponsive state
Looking back at the history - these pool refreshes and status calls were the only traffic before the application became unresponsive: curl -X PUT -H "Authorization: Basic Y2FuZGxlcGluX2FkbWluOmNhbmRsZXBpbl9hZG1pbg==" -H "Cache-Control: no-cache" "http://candlepin06.web.dev.int.devlab.redhat.com:8080/candlepin/owners/7388484/subscriptions?auto_create_owner=true" curl -X PUT -H "Authorization: Basic Y2FuZGxlcGluX2FkbWluOmNhbmRsZXBpbl9hZG1pbg==" -H "Cache-Control: no-cache" "http://candlepin06.web.dev.int.devlab.redhat.com:8080/candlepin/owners/6302666/subscriptions?auto_create_owner=true"
Created attachment 1153034 [details] Thread dump