----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/29901/#review68164 -----------------------------------------------------------
src/test/python/apache/aurora/executor/common/test_resource_manager_integration.py <https://reviews.apache.org/r/29901/#comment112326> Is this timeout necessary? In our experience Java-side, flaky tests that get fixed with magic timeouts tend to break again later if the CI environment becomes loaded. Is there a way you could write this in a more deterministic fashion, e.g.: ``` 1. write >limit bytes 2. force the resource manager to poll (via a simulated clock or directly calling a poll method visible for testing) 3. check the status ``` - Kevin Sweeney On Jan. 14, 2015, 3:26 p.m., Brian Wickman wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/29901/ > ----------------------------------------------------------- > > (Updated Jan. 14, 2015, 3:26 p.m.) > > > Review request for Aurora and Kevin Sweeney. > > > Repository: aurora > > > Description > ------- > > De-flakify ResourceManager integration test. > > > Diffs > ----- > > src/main/python/apache/aurora/executor/common/resource_manager.py > bf69e7e60ae4dd857243d8b040a70daca008ea28 > > src/test/python/apache/aurora/executor/common/test_resource_manager_integration.py > e3c766f9b7bf989c0790281231957947d2cda52a > > Diff: https://reviews.apache.org/r/29901/diff/ > > > Testing > ------- > > ./pants goal test --no-test-pytest-fast > src/test/python/apache/aurora/executor/common:: > > > Thanks, > > Brian Wickman > >
