> On Jan. 22, 2018, 8:51 p.m., Gaston Kleiman wrote: > > src/tests/api_tests.cpp > > Lines 1011-1013 (patched) > > <https://reviews.apache.org/r/65044/diff/5/?file=1943283#file1943283line1012> > > > > This doesn't seem to be necessary in most other tests, why exaclty do > > we need it here? > > Jan Schlicht wrote: > The allocation interval is set to 1000s in the test fixture to avoid a > race. This is a problem in this particular test case as it is the only one in > this fixture that runs a framework and needs to get offers.
I agree that this isn't very intuitive, I've found a better solution by settling the clock before advancing the clock to make sure that a framework is registered. Then it also won't run in authentication timeouts. > On Jan. 22, 2018, 8:51 p.m., Gaston Kleiman wrote: > > src/tests/api_tests.cpp > > Lines 6398 (patched) > > <https://reviews.apache.org/r/65044/diff/5/?file=1943283#file1943283line6399> > > > > Wouldn't this expectation fail if the master received a terminal > > operation status update before doing the `GET_OPERATIONS` call? > > > > We probably want to drop all `UpdateOperationStatusMessage` protobufs > > from the agent to the salve. In that case the operation would still show up with `OPERATION_FINISHED` state (as it's tested in the follow up RR). But as we control the behavior of the resource provider in this test, the operation status won't be terminal at this point, as we didn't do anything in response to the `applyOperation` call. - Jan ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/65044/#review195927 ----------------------------------------------------------- On Jan. 22, 2018, 3:02 p.m., Jan Schlicht wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/65044/ > ----------------------------------------------------------- > > (Updated Jan. 22, 2018, 3:02 p.m.) > > > Review request for mesos, Benjamin Bannier, Gaston Kleiman, Greg Mann, and > Jie Yu. > > > Bugs: MESOS-8424 > https://issues.apache.org/jira/browse/MESOS-8424 > > > Repository: mesos > > > Description > ------- > > The 'GET_OPERATIONS' call lists all operations known to master or agent. > > > Diffs > ----- > > include/mesos/agent/agent.proto 315820041990d28e5cdae71c8385d4551d56558c > include/mesos/master/master.proto 3e34634f3b864222d07374936c0d9a18269c2fbd > include/mesos/v1/agent/agent.proto 9e8b49defb83ffd4dd240ebb3a09b2dd3610ab2a > include/mesos/v1/master/master.proto > 6759c3004e7e04c0044360b0c54f4438c2f6ec8a > src/master/http.cpp c489b6f525f157811549b2cc84addd9d85e87990 > src/master/master.hpp 05136781b6b1539f37c283e8127e4bafb187a0d1 > src/master/validation.cpp f0b86775b7919ba6aa4a73038edb78a0adca68f4 > src/slave/http.hpp 1619bb77e3e7d5a6d387e4d4bb071d83ce914967 > src/slave/http.cpp 77e711ceeb0e2613d629b5e21fd686f85dfad11a > src/slave/validation.cpp 0c2ccda177734cf3c47c0346ed34d20d58e7d932 > src/tests/api_tests.cpp bdacc30be4dc8656a41a0c47d0e350d48e59ad94 > > > Diff: https://reviews.apache.org/r/65044/diff/5/ > > > Testing > ------- > > make check > > > Thanks, > > Jan Schlicht > >
