----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/50214/#review143589 -----------------------------------------------------------
Fix it, then Ship it! src/slave/containerizer/mesos/launch.cpp (line 220) <https://reviews.apache.org/r/50214/#comment209414> I'd do the following: ``` Try<Subprocess> s = Error("Not launched"); if (parse->shell()) { s = subprocess(...); } else { s = subprocess(...); } if (s.isError()) { ... } Option<int> status = s->status().await(); ... ``` src/slave/containerizer/mesos/launch.cpp (line 250) <https://reviews.apache.org/r/50214/#comment209415> Not yours, but can you follow up with a patch to use return EXIT_FAILURE instead? - Jie Yu On July 26, 2016, 3:19 a.m., Gilbert Song wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/50214/ > ----------------------------------------------------------- > > (Updated July 26, 2016, 3:19 a.m.) > > > Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen. > > > Bugs: MESOS-5388 > https://issues.apache.org/jira/browse/MESOS-5388 > > > Repository: mesos > > > Description > ------- > > Currently all pre_exec_commands are executed as shell commands > in Mesos Launch. It is not safe because arbitrary shell command > may be included in some user facing api (e.g., container_path). > We should execute those command as a subprocess to prevent > arbitrary shell command injection. > > > Diffs > ----- > > src/slave/containerizer/mesos/launch.cpp > 51f0c110ff0c414837fd69db81047979a0093388 > > Diff: https://reviews.apache.org/r/50214/diff/ > > > Testing > ------- > > make check > > sudo ./bin/mesos-tests.sh > > > Thanks, > > Gilbert Song > >
