[ https://forge.continuent.org/jira/browse/SEQUOIA-633?page=all ] Damian Arregui closed SEQUOIA-633: ----------------------------------
> Potential deadlock when a transaction is being aborted and a task completes > its execution simultaneously > -------------------------------------------------------------------------------------------------------- > > Key: SEQUOIA-633 > URL: https://forge.continuent.org/jira/browse/SEQUOIA-633 > Project: Sequoia > Type: Bug > Versions: Sequoia 2.8.2 > Reporter: Damian Arregui > Assignee: Damian Arregui > Priority: Critical > Fix For: Sequoia 2.9 > > > Deadlock occurs because locks on a particular backend task queue and the > BackendTaskQueues object are not always acquired in the same order. > When executing a distributed abort, MulticastRequestAdapterThread LOCKS a > particular backend task queue (in > BTQ.abortAllQueriesEvenRunninInTransaction()) and WAITS FOR the > BackendTaskQueues object (in BTQ.completeEntryExecution()). > That objet is held by a BackendWorkerThread which WAITS FOR the same > particular backend task queue (in BTQ.completeEntryExecution()). > Locks should be acquired always in the same order. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://forge.continuent.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira _______________________________________________ Sequoia mailing list [email protected] https://forge.continuent.org/mailman/listinfo/sequoia
