----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/25373/#review52611 -----------------------------------------------------------
src/tests/fault_tolerance_tests.cpp <https://reviews.apache.org/r/25373/#comment91460> s/executor ,/executor,/ src/tests/fault_tolerance_tests.cpp <https://reviews.apache.org/r/25373/#comment91461> s/the the/the/ src/tests/fault_tolerance_tests.cpp <https://reviews.apache.org/r/25373/#comment91464> What is the guarantee here that the 2nd task is pending on the slave at this point? src/tests/fault_tolerance_tests.cpp <https://reviews.apache.org/r/25373/#comment91465> Instead of having a mock gc executor and launching 2 tasks and pausing the gc's unschedule, can we just do: launch one task. drop _runTask() dispatch. force re-registration. is that not enough? - Vinod Kone On Sept. 5, 2014, 3:10 a.m., Ben Mahler wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/25373/ > ----------------------------------------------------------- > > (Updated Sept. 5, 2014, 3:10 a.m.) > > > Review request for mesos and Vinod Kone. > > > Bugs: MESOS-1715 > https://issues.apache.org/jira/browse/MESOS-1715 > > > Repository: mesos-git > > > Description > ------- > > This uses a MockGarbageCollector and a detector to force the slave to > re-register with a pending task. > > > Diffs > ----- > > src/tests/fault_tolerance_tests.cpp > b0e74b94db8a68cd676569d42b2ccd120f12b9d3 > > Diff: https://reviews.apache.org/r/25373/diff/ > > > Testing > ------- > > make check > > > Thanks, > > Ben Mahler > >
