----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15872/#review29496 -----------------------------------------------------------
Ship it! src/exec/exec.cpp <https://reviews.apache.org/r/15872/#comment56802> There "may be at most one"? src/sched/sched.cpp <https://reviews.apache.org/r/15872/#comment56803> Want to fix this too? ;) src/sched/sched.cpp <https://reviews.apache.org/r/15872/#comment56804> ExecutorProcess? src/sched/sched.cpp <https://reviews.apache.org/r/15872/#comment56805> ExecutorProcess? src/sched/sched.cpp <https://reviews.apache.org/r/15872/#comment56806> At most one? - Benjamin Hindman On Nov. 27, 2013, 4:12 a.m., Ben Mahler wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/15872/ > ----------------------------------------------------------- > > (Updated Nov. 27, 2013, 4:12 a.m.) > > > Review request for mesos, Benjamin Hindman and Vinod Kone. > > > Bugs: MESOS-851 > https://issues.apache.org/jira/browse/MESOS-851 > > > Repository: mesos-git > > > Description > ------- > > See MESOS-851. > > Now once a call to SchedulerDriver::abort() is issued, we begin ignoring > messages that were queued in the driver. Previously, we would have to process > the entire queue of messages, which can result in many calls to the > Scheduler, despite abort() already having been called. > > This ensures that once abort() is called, we ignore queued messages while > still processing queued requests *from* the scheduler. > > The same change here was applied to the ExecutorDriver. > > > Diffs > ----- > > src/exec/exec.cpp a58203b1ce85af9711dc207c34e7bc673c5441b1 > src/sched/sched.cpp af57f284b715e1498fd4b2a317706bc33792dec3 > > Diff: https://reviews.apache.org/r/15872/diff/ > > > Testing > ------- > > make check > > > Thanks, > > Ben Mahler > >
