----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/21329/#review42751 -----------------------------------------------------------
Ship it! src/launcher/executor.cpp <https://reviews.apache.org/r/21329/#comment76627> Alphabetize please. src/launcher/executor.cpp <https://reviews.apache.org/r/21329/#comment76629> Extra newline please. src/tests/slave_tests.cpp <https://reviews.apache.org/r/21329/#comment76637> You'll want to use 'test::flags.build_dir' instead of this object. - Benjamin Hindman On May 12, 2014, 8:59 p.m., Niklas Nielsen wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/21329/ > ----------------------------------------------------------- > > (Updated May 12, 2014, 8:59 p.m.) > > > Review request for mesos, Benjamin Hindman, Jason Dusek, and Till Toenshoff. > > > Repository: mesos-git > > > Description > ------- > > It will be a common use case that external containerizer programs will have > to be able to control the command being carried out when the only the > task.command or container image is present. > The command executor will in its current state try to run the command in the > context of the slave and not the container. > > So, you could imagine the command executor being used by: > 1) Replace the command to run when the slave sends a runTask message > 2) Prefix a command to wrap call in when the slave sends a runTask message. > > 1) have been implemented and tested in this patch, alongside making the > command-executor "cli-ized" with help messages. > > > Diffs > ----- > > src/launcher/executor.cpp 312606b > src/tests/containerizer.hpp 8e21bd1 > src/tests/containerizer.cpp 7dacab5 > src/tests/slave_tests.cpp 458356d > > Diff: https://reviews.apache.org/r/21329/diff/ > > > Testing > ------- > > make check with new MesosExecutorWithOverride test. > > > Thanks, > > Niklas Nielsen > >
