> On March 3, 2014, 6:16 p.m., Maxim Khutornenko wrote: > > src/main/java/org/apache/aurora/scheduler/SchedulerLifecycle.java, line 236 > > <https://reviews.apache.org/r/18713/diff/1/?file=509104#file509104line236> > > > > Not sure how shutdownRegistry.addAction() fits into this specific > > Closure. It there anything forcing it here? If not shouldn't be one level > > up within the constructor body? > > Suman Karumuri wrote: > +1
+1, I think this should happen right after the state machine is created. - Kevin ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/18713/#review36077 ----------------------------------------------------------- On March 3, 2014, 3:31 p.m., Bill Farner wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/18713/ > ----------------------------------------------------------- > > (Updated March 3, 2014, 3:31 p.m.) > > > Review request for Aurora, Suman Karumuri and Maxim Khutornenko. > > > Repository: aurora > > > Description > ------- > > AFAICT there currently isn't wiring for this, so application exit not > initiated by scheduler lifecycle will not gracefully tear down components > handled by SchedulerLifecycle. > > > Diffs > ----- > > src/main/java/org/apache/aurora/scheduler/SchedulerLifecycle.java > 1cdbff62b5348db4565388651032bf88533d9d91 > src/test/java/org/apache/aurora/scheduler/SchedulerLifecycleTest.java > 1b1e948d95387fe6aab0a78c2b1b1744b990ba55 > > Diff: https://reviews.apache.org/r/18713/diff/ > > > Testing > ------- > > ./gradlew build > > > Thanks, > > Bill Farner > >
