> On April 3, 2015, 3:46 p.m., Benjamin Hindman wrote: > > src/slave/containerizer/docker.cpp, line 1230 > > <https://reviews.apache.org/r/32797/diff/1/?file=914221#file914221line1230> > > > > Why kill the executor before doing Docker::stop? Can you comment here > > why you do it in this order versus the other order and the ramifications > > that has?
This is because we're waiting on the executor to finish (os::reaped(executorPid)) in the container->status future, and if we don't kill the executor first the later container->status call will just hang. I can leave a comment about this too. - Timothy ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/32797/#review78785 ----------------------------------------------------------- On April 2, 2015, 11:38 p.m., Timothy Chen wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/32797/ > ----------------------------------------------------------- > > (Updated April 2, 2015, 11:38 p.m.) > > > Review request for mesos, Benjamin Hindman, Bernd Mathiske, and Till > Toenshoff. > > > Bugs: MESOS-2583 > https://issues.apache.org/jira/browse/MESOS-2583 > > > Repository: mesos > > > Description > ------- > > Kill the executor when docker container is destroyed. > > > Diffs > ----- > > src/slave/containerizer/docker.hpp b7bf54ac65d6c61622e485ac253513eaac2e4f88 > src/slave/containerizer/docker.cpp e83b912c707a3f2687b09a647a9ed248a940ea97 > > Diff: https://reviews.apache.org/r/32797/diff/ > > > Testing > ------- > > make check > > > Thanks, > > Timothy Chen > >
