> On June 3, 2013, 11:27 p.m., Ben Mahler wrote: > > src/tests/fault_tolerance_tests.cpp, lines 1341-1346 > > <https://reviews.apache.org/r/11543/diff/2/?file=300136#file300136line1341> > > > > Can you comment that these DROPs are to ensure the terminated Executor > > sticks around in the slave, awaiting update acknowledgments? > > > > In particular why are both of these needed? > > Vinod Kone wrote: > actually the second one is not needed. killed it.
pressed submit too soon. actually dropping it is not required, but waiting on it is, to ensure slave re-registration actually happens after executor exit is known to the slave. - Vinod ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/11543/#review21367 ----------------------------------------------------------- On June 3, 2013, 10:04 p.m., Vinod Kone wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/11543/ > ----------------------------------------------------------- > > (Updated June 3, 2013, 10:04 p.m.) > > > Review request for mesos and Benjamin Hindman. > > > Description > ------- > > See summary. > > > This addresses bug MESOS-481. > https://issues.apache.org/jira/browse/MESOS-481 > > > Diffs > ----- > > src/slave/slave.cpp e905ab3cd965002592c1f70c85ea3378ac8982b0 > src/tests/fault_tolerance_tests.cpp > c8fad7c02abcafa8ee9351a8b40684dba7f2c711 > > Diff: https://reviews.apache.org/r/11543/diff/ > > > Testing > ------- > > make check > > ./bin/mesos-tests.sh --gtest_filter="*SlaveReregisterTerminatedExecutor*" > --verbose --gtest_repeat=1000 --gtest_break_on_failure > > > Thanks, > > Vinod Kone > >
