----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/65077/#review195219 -----------------------------------------------------------
src/tests/containerizer/composing_containerizer_tests.cpp Lines 363-371 (patched) <https://reviews.apache.org/r/65077/#comment274396> This part can also passed without recovering above, right? Maybe we do not want to verify the underlying wait() is called or not under the composing containerizer `recover()`. Since this is a regression test, could we have an end-to-end test (restart the agent and kill a task, verify the task with correct status)? - Gilbert Song On Jan. 10, 2018, 10:54 a.m., Andrei Budnik wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/65077/ > ----------------------------------------------------------- > > (Updated Jan. 10, 2018, 10:54 a.m.) > > > Review request for mesos, Alexander Rukletsov, Gilbert Song, and Jie Yu. > > > Bugs: MESOS-8391 > https://issues.apache.org/jira/browse/MESOS-8391 > > > Repository: mesos > > > Description > ------- > > This test verifies that `wait()` called on composing containerizer for > a container that has been recovered, returns a future set to `READY` > iff the container is terminated. > > > Diffs > ----- > > src/tests/containerizer/composing_containerizer_tests.cpp > 7c22f162b128c3fdf8d4b20cac73fdf442449d79 > > > Diff: https://reviews.apache.org/r/65077/diff/1/ > > > Testing > ------- > > ./src/mesos-tests --gtest_filter=ComposingContainerizerTest.WaitAfterRecovery > --gtest_break_on_failure --gtest_repeat=100 --verbose > > > Thanks, > > Andrei Budnik > >
