DOKK / manpages / debian 10 / charliecloud / ch-run.1.en
CH-RUN(1) Charliecloud CH-RUN(1)

ch-run - Run a command in a Charliecloud container

$ ch-run [OPTION...] NEWROOT CMD [ARG...]


Run command CMD in a Charliecloud container using the flattened and unpacked image directory located at NEWROOT.

mount SRC at guest DST (default /mnt/0, /mnt/1, etc.)
initial working directory in container
run as group GID within container
use the same container (namespaces) as peer ch-run invocations
join the namespaces of an existing process
number of ch-run peers (implies --join; default: see below)
label for ch-run peer group (implies --join; default: see below)
do not bind-mount your home directory (by default, your home directory is mounted at /home/$USER in the container)
use container-private /tmp (by default, /tmp is shared with the host)
run as user UID within container
be more verbose (debug if repeated)
mount image read-write (by default, the image is mounted read-only)
-?, --help
print help and exit
print a short usage message and exit
print version and exit



In addition to any directories specified by the user with --bind, ch-run has standard host files and directories that are bind-mounted in as well.

The following host files and directories are bind-mounted at the same location in the container. These cannot be disabled.

  • /dev
  • /etc/passwd
  • /etc/group
  • /etc/hosts
  • /etc/resolv.conf
  • /proc
  • /sys



Three additional bind mounts can be disabled by the user:

  • Your home directory (i.e., $HOME) is mounted at guest /home/$USER by default. This is accomplished by mounting a new tmpfs at /home, which hides any image content under that path. If --no-home is specified, neither of these things happens and the image’s /home is exposed unaltered.
  • /tmp is shared with the host by default. If --private-tmp is specified, a new tmpfs is mounted on the guest’s /tmp instead.
  • If file /usr/bin/ch-ssh is present in the image, it is over-mounted with the ch-ssh binary in the same directory as ch-run.



By default, different ch-run invocations use different user and mount namespaces (i.e., different containers). While this has no impact on sharing most resources between invocations, there are a few important exceptions. These include:

1.
ptrace(2), used by debuggers and related tools. One can attach a debugger to processes in descendant namespaces, but not sibling namespaces. The practical effect of this is that (without --join), you can’t run a command with ch-run and then attach to it with a debugger also run with ch-run.
2.
Cross-memory attach (CMA) is used by cooperating processes to communicate by simply reading and writing one another’s memory. This is also not permitted between sibling namespaces. This affects various MPI implementations that use CMA to pass messages between ranks on the same node, because it’s faster than traditional shared memory.

--join is designed to address this by placing related ch-run commands (the “peer group”) in the same container. This is done by one of the peers creating the namespaces with unshare(2) and the others joining with setns(2).

To do so, we need to know the number of peers and a name for the group. These are specified by additional arguments that can (hopefully) be left at default values in most cases:

  • --join-ct sets the number of peers. The default is the value of the first of the following environment variables that is defined: OMPI_COMM_WORLD_LOCAL_SIZE, SLURM_STEP_TASKS_PER_NODE, SLURM_CPUS_ON_NODE.
  • --join-tag sets the tag that names the peer group. The default is environment variable SLURM_STEP_ID, if defined; otherwise, the PID of ch-run’s parent. Tags can be re-used for peer groups that start at different times, i.e., once all peer ch-run have replaced themselves with the user command, the tag can be re-used.

Caveats:

  • One cannot currently add peers after the fact, for example, if one decides to start a debugger after the fact. (This is only required for code with bugs and is thus an unusual use case.)
  • ch-run instances race. The winner of this race sets up the namespaces, and the other peers use the winner to find the namespaces to join. Therefore, if the user command of the winner exits, any remaining peers will not be able to join the namespaces, even if they are still active. There is currently no general way to specify which ch-run should be the winner.
  • If --join-ct is too high, the winning ch-run’s user command exits before all peers join, or ch-run itself crashes, IPC resources such as semaphores and shared memory segments will be leaked. These appear as files in /dev/shm/ and can be removed with rm(1).
  • Many of the arguments given to the race losers, such as the image path and --bind, will be ignored in favor of what was given to the winner.

ch-run generally tries to leave environment variables unchanged, but in some cases, guests can be significantly broken unless environment variables are tweaked. This section lists those changes.

  • $HOME: If the path to your home directory is not /home/$USER on the host, then an inherited $HOME will be incorrect inside the guest. This confuses some software, such as Spack.

    Thus, we change $HOME to /home/$USER, unless --no-home is specified, in which case it is left unchanged.

  • $PATH: Newer Linux distributions replace some root-level directories, such as /bin, with symlinks to their counterparts in /usr.

    Some of these distributions (e.g., Fedora 24) have also dropped /bin from the default $PATH. This is a problem when the guest OS does not have a merged /usr (e.g., Debian 8 “Jessie”). Thus, we add /bin to $PATH if it’s not already present.

    Further reading:

  • The case for the /usr Merge
  • Fedora
  • Debian




Run the command echo hello inside a Charliecloud container using the unpacked image at /data/foo:

$ ch-run /data/foo -- echo hello
hello


Run an MPI job that can use CMA to communicate:

$ srun ch-run --join /data/foo -- bar


If Charliecloud was obtained from your Linux distribution, use your distribution’s bug reporting procedures.

Otherwise, report bugs to: <https://github.com/hpc/charliecloud/issues>

charliecloud(1)

Full documentation at: <https://hpc.github.io/charliecloud>

Reid Priedhorsky, Tim Randles, and others

2014–2018, Los Alamos National Security, LLC

2018-12-14 19:09 Coordinated Universal Time