----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/51122/#review145832 -----------------------------------------------------------
src/slave/posix_signalhandler.hpp (lines 39 - 47) <https://reviews.apache.org/r/51122/#comment212188> This note seems to suggest that we should only set up the signal handler from main.cpp? If the behavior is undefined, seems we should CHECK against this? - Benjamin Mahler On Aug. 16, 2016, 1:34 a.m., Joseph Wu wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/51122/ > ----------------------------------------------------------- > > (Updated Aug. 16, 2016, 1:34 a.m.) > > > Review request for mesos, Benjamin Mahler, Greg Mann, Artem Harutyunyan, and > Vinod Kone. > > > Repository: mesos > > > Description > ------- > > This leak was introduced on purpose in: > https://reviews.apache.org/r/34016/ > > When we run tests in repetition, each agent will create a new signal > handler and leak the old one. We can either delete the old signal > handler, or guard this function with a `Once`. > > > Diffs > ----- > > src/slave/posix_signalhandler.hpp 4a5411793e55f912c6e8ae72f9a8042d3a6195b5 > > Diff: https://reviews.apache.org/r/51122/diff/ > > > Testing > ------- > > make check (OSX) > > Found this leak via code inspection after valgrind. > > > Thanks, > > Joseph Wu > >
