----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/63279/#review190562 -----------------------------------------------------------
Ship it! src/tests/check_tests.cpp Lines 235 (patched) <https://reviews.apache.org/r/63279/#comment268040> cpu is not throttable on windows?! - Alexander Rukletsov On Nov. 2, 2017, 8:40 p.m., Andrew Schwartzmeyer wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/63279/ > ----------------------------------------------------------- > > (Updated Nov. 2, 2017, 8:40 p.m.) > > > Review request for mesos, Aaron Wood, Akash Gupta, Alexander Rukletsov, Jeff > Coffler, Gaston Kleiman, Jie Yu, John Kordich, Joseph Wu, Li Li, and Michael > Park. > > > Bugs: MESOS-6690 > https://issues.apache.org/jira/browse/MESOS-6690 > > > Repository: mesos > > > Description > ------- > > Unfortunately, due to PowerShell's resource usage as a .NET program, on > Windows 32 MB is not enough memory to run these tests. One instance of > PowerShell takes > 128 MB, and two instances take > 256 MB. > Realistically, the safe minimum is 512 MB of memory. > > > Diffs > ----- > > src/tests/check_tests.cpp d3ffc0b0204d021f1ed734ece189d0e3a3fd8844 > > > Diff: https://reviews.apache.org/r/63279/diff/2/ > > > Testing > ------- > > Built and ran `mesos-tests.exe` on Windows repeatedly, verified all tests > still pass consistently (likewise for `stout` and `libprocess` tests. > > ~~Currently verifying no breaks on Linux.~~ Built and ran tests on Linux. > > NOTE: There are more check tests that are currently disabled for Windows, > that I think should be enabled, but did immediately work for me, so I've left > them disabled to unblock myself. Similarly, I would ideally like to port the > balloon example framework, and use that to prove the effectiveness of the job > object memory hard-cap. Having not yet ported it though, I manually verified > the effectiveness of the new isolators by launching test CPU and memory test > tasks on a deployed cluster my these changes (and the fact that PowerShell > OOM'ed was a nice verification too). > > > Thanks, > > Andrew Schwartzmeyer > >