> On May 30, 2014, 5:32 p.m., Vinod Kone wrote: > > What is the bug and what is the fix here? > > Dominic Hamon wrote: > we weren't waiting for anything before trying to hit the endpoint and > exiting the test. As such, the scheduler wasn't registering. By adding a > slave and waiting for offers, we ensure that everything is running before > testing.
Is the scheduler registering important for this test? If not, then why not just do .Times(AtMost(1)) on registered expectation. OTH, if registration is necessary, why not explicitly wait for registration instead of starting a slave and waiting for offers as a proxy for registration? - Vinod ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/22089/#review44384 ----------------------------------------------------------- On May 30, 2014, 5:08 p.m., Dominic Hamon wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/22089/ > ----------------------------------------------------------- > > (Updated May 30, 2014, 5:08 p.m.) > > > Review request for mesos, Vinod Kone and Jiang Yan Xu. > > > Bugs: MESOS-1439 > https://issues.apache.org/jira/browse/MESOS-1439 > > > Repository: mesos-git > > > Description > ------- > > see summary > > > Diffs > ----- > > src/tests/scheduler_tests.cpp 721d1cf74d980649f43dfb9e8778036f1b7780a6 > > Diff: https://reviews.apache.org/r/22089/diff/ > > > Testing > ------- > > make check x 400+ > > > Thanks, > > Dominic Hamon > >
