> On Oct. 31, 2013, 5:59 p.m., Benjamin Hindman wrote: > > src/sched/sched.cpp, line 288 > > <https://reviews.apache.org/r/15138/diff/1/?file=375163#file375163line288> > > > > If this defer never gets invoked now the authenticatee will never get > > deleted causing a memory leak, hence the use of Owned. Also, I don't > > understand how this causes a deadlock? Finally, to keep people from > > rewrapping this with Owned you would probably want to a big comment. ;) > > Vinod Kone wrote: > Also not sure about the leak because afaict the only reason the defer > wouldn't be called is when the process is exiting.
Exactly. If the SchedulerProcess is exiting then we won't execute the defer which means the pointer will not get deleted, thus a memory leak. - Benjamin ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15138/#review27929 ----------------------------------------------------------- On Oct. 31, 2013, 6:24 p.m., Vinod Kone wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/15138/ > ----------------------------------------------------------- > > (Updated Oct. 31, 2013, 6:24 p.m.) > > > Review request for mesos, Benjamin Hindman and Ben Mahler. > > > Bugs: MESOS-787 > https://issues.apache.org/jira/browse/MESOS-787 > > > Repository: mesos-git > > > Description > ------- > > See bug for details. > > > Diffs > ----- > > src/sched/sched.cpp 3049096a7aeaea6c1063a1bbcef119517daa236c > > Diff: https://reviews.apache.org/r/15138/diff/ > > > Testing > ------- > > make check > > > Thanks, > > Vinod Kone > >
