----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/61570/#review183103 -----------------------------------------------------------
src/slave/containerizer/mesos/containerizer.cpp Lines 2463 (patched) <https://reviews.apache.org/r/61570/#comment259105> I don't like this semantics. The signal here can be any signal, not necessarily a SIGKILL/SIGTERM. For instance, SIGUSR1, SIGALARM, etc., which might not be terminal signal. Any reason why we want to destroy the container here? - Jie Yu On Aug. 14, 2017, 11:09 p.m., Anand Mazumdar wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/61570/ > ----------------------------------------------------------- > > (Updated Aug. 14, 2017, 11:09 p.m.) > > > Review request for mesos, Jie Yu and Vinod Kone. > > > Bugs: MESOS-7879 > https://issues.apache.org/jira/browse/MESOS-7879 > > > Repository: mesos > > > Description > ------- > > This would be used by now on for killing a container by sending > a signal to it similar to the linux equivalent `kill()` system call. > > > Diffs > ----- > > src/slave/containerizer/containerizer.hpp > 0954ed6175e4c8f963bf371e54e0f9ffe7bc9c1c > src/slave/containerizer/mesos/containerizer.hpp > fd586306f7a6d3d2cdb58481f0ef4906de8d0f88 > src/slave/containerizer/mesos/containerizer.cpp > ff192bb085f3554ad1b1f20fb73bf827bf04ef13 > > > Diff: https://reviews.apache.org/r/61570/diff/3/ > > > Testing > ------- > > make check > > > Thanks, > > Anand Mazumdar > >
