[ https://issues.apache.org/jira/browse/IMPALA-6194?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16703905#comment-16703905 ]
Lars Volker commented on IMPALA-6194: ------------------------------------- Are I/O threads the only ones that can get stuck? The idea of the Jira was to introduce a systematic way of ensuring that queries cannot get stuck, e.g. by making it much harder to wait on a CV that doesn't get signaled during cancellation. I'm in favor of tracking the general problem here. > Ensure all fragment instances notice cancellation > ------------------------------------------------- > > Key: IMPALA-6194 > URL: https://issues.apache.org/jira/browse/IMPALA-6194 > Project: IMPALA > Issue Type: Improvement > Components: Backend > Reporter: Lars Volker > Priority: Critical > Labels: observability, supportability > > Currently queries can get stuck in an uncancellable state, e.g. when blocking > on function calls or condition variables without periodically checking for > cancellation. We should eliminate all those calls and make sure we don't > re-introduce such issues. One option would be a watchdog to check that each > fragment instance regularly calls RETURN_IF_CANCEL. -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-all-unsubscr...@impala.apache.org For additional commands, e-mail: issues-all-h...@impala.apache.org