> On Feb. 27, 2014, 5:16 p.m., Vinod Kone wrote: > > src/launcher/executor.cpp, lines 230-236 > > <https://reviews.apache.org/r/18597/diff/1/?file=506548#file506548line230> > > > > Since the slave is already doing the signal escalation, what is the > > benefit of doing it here? > > > > Niklas Nielsen wrote: > If we can change the SIGKILL to SIGTERM here and let container->destroy() > take care of cleaning up; then no need (but changing the signal sent here). I > assumed you ran into problems with SIGTERM only, but that might have been > before the containerizer changes? > > Vinod Kone wrote: > I think with the latest containerizer refactor and an improved killtree > we should do the right thing. > > If all we want to do is s/SIGKILL/SIGTERM/ that actually sounds good to > me. Ofcourse we have to make sure all our unit tests pass :)
I have run makc check with SIGTERM on both Mac and Linux; works fine :) I'll drop these RR's and file a single one with s/SIGTERM/SIGKILL/ - Niklas ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/18597/#review35739 ----------------------------------------------------------- On Feb. 27, 2014, 4:37 p.m., Niklas Nielsen wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/18597/ > ----------------------------------------------------------- > > (Updated Feb. 27, 2014, 4:37 p.m.) > > > Review request for mesos. > > > Bugs: MESOS-1031 > https://issues.apache.org/jira/browse/MESOS-1031 > > > Repository: mesos-git > > > Description > ------- > > This patch makes command executor shutdown a bit more graceful > (than sending SIGKILL upfront) by adding signal escalation. > Signal escalation checks for liveness of process trees after SIGTERM > has been sent. If all processes are dead before the given gracePeriod, > shutdown() returns. If not, SIGKILL will be sent to all pids in > the process trees. gracePeriod is set to 3 seconds, in order to issue > SIGKILL before shutdownTimeout in the slave is triggered > (EXECUTOR_SHUTDOWN_GRACE_PERIOD is 5 seconds). > > > Diffs > ----- > > src/launcher/executor.cpp e30d77a > > Diff: https://reviews.apache.org/r/18597/diff/ > > > Testing > ------- > > Functional testing and make check. > > > Thanks, > > Niklas Nielsen > >
