> On March 7, 2017, 7:28 p.m., Jiang Yan Xu wrote: > > So I was thinking about this: the code here makes sure the OS has the > > minimum set of subsystems required by this test suite. If the system > > doesn't meet the requirement, we do want to terminate here. On the other > > hand, although mounting more subsystems isn't harmful, if we mount > > everything that comes out of `cgroups::subsystems()`, we should still check > > whether that list is a superset of the minimumally required set. At a > > result we still have to add items here every time a new subsystem is added > > so we are not saving much? > > > > Thoughts?
I think you are reading too much into this. Is this really a contract about what subsystems the tests require? I'd argue that it is just an arbitrary selection from history. Mounting everything is robust and requires no future maintenance. Tests that need to verify specific subsystems are present can do that. - James ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/56733/#review168173 ----------------------------------------------------------- On March 6, 2017, 7:34 p.m., James Peach wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/56733/ > ----------------------------------------------------------- > > (Updated March 6, 2017, 7:34 p.m.) > > > Review request for mesos, Mesos Reviewbot and Jiang Yan Xu. > > > Bugs: MESOS-7049 > https://issues.apache.org/jira/browse/MESOS-7049 > > > Repository: mesos > > > Description > ------- > > Rather than a hard-coded list of subsystems, just mount everything we > find that is supported. > > > Diffs > ----- > > src/tests/mesos.cpp 6a96fa51dfc2a62063c3154b256bdac707b009bb > > > Diff: https://reviews.apache.org/r/56733/diff/2/ > > > Testing > ------- > > sudo make check (Fedora 25) > > > Thanks, > > James Peach > >
