Hi Bjoern, I see that you committed the patch and also closed the issue. It is good practice to let the reporter close the issue, this gives them a change to verify it is ok.
Usually I mark the issue as fixed/resolved and then ask the reporter to close it. Anyway, no harm done and good to see you active! 2014-06-08 22:58 GMT+02:00 Bjoern Petri (JIRA) <[email protected]>: > > [ > https://issues.apache.org/jira/browse/CELIX-114?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel > ] > > Bjoern Petri resolved CELIX-114. > -------------------------------- > > Resolution: Fixed > Assignee: Bjoern Petri > > patch committed for potential deadlock > > > Potential deadlock in log_service bundle during stop > > ---------------------------------------------------- > > > > Key: CELIX-114 > > URL: https://issues.apache.org/jira/browse/CELIX-114 > > Project: Celix > > Issue Type: Bug > > Reporter: Gabriele Ricciardi > > Assignee: Bjoern Petri > > Attachments: log_service_deadlock_patch.tgz > > > > > > In a non-sistematic way, the log_service bundle gets stuck during the > framework shutdown phase. In particular, after analyzing the core file > generated by the application abort, seems that a deadlock condition exists > among the removeListener function (blocked on the apr_thread_join) and the > logListener thread (waiting on a never-again signalled condition variable). > > Attached a patch solving the issue > > > > -- > This message was sent by Atlassian JIRA > (v6.2#6252) > -- Met vriendelijke groet, Alexander Broekhuis
