Ps command may return exit 1 when no processes are shown

[sirn]

On some systems such as FreeBSD, running ps without an assigned ttys, without login shell and without displaying header may cause ps command to have no output and exit 1 (e.g. via ssh user@host ps -o pid= or as part of Ansible script). When this happens, shellingham may fail with subprocess.CalledProcessError:

In particular, I got this error when I tried to run pipenv over Ansible on FreeBSD:

Traceback (most recent call last):
  File "/usr/local/bin/pipenv", line 7, in <module>
    from pipenv import cli
  File "/usr/local/lib/python3.6/site-packages/pipenv/__init__.py", line 23, in <module>
    from .cli import cli
  File "/usr/local/lib/python3.6/site-packages/pipenv/cli/__init__.py", line 3, in <module>
    from .command import cli
  File "/usr/local/lib/python3.6/site-packages/pipenv/cli/command.py", line 7, in <module>
    import crayons
  File "/usr/local/lib/python3.6/site-packages/pipenv/patched/crayons.py", line 49, in <module>
    is_powershell = "powershell" in shellingham.detect_shell()[0]
  File "/usr/local/lib/python3.6/site-packages/pipenv/vendor/shellingham/__init__.py", line 22, in detect_shell
    shell = get_shell(pid, max_depth=max_depth)
  File "/usr/local/lib/python3.6/site-packages/pipenv/vendor/shellingham/posix/__init__.py", line 54, in get_shell
    mapping = _get_process_mapping()
  File "/usr/local/lib/python3.6/site-packages/pipenv/vendor/shellingham/posix/__init__.py", line 15, in _get_process_mapping
    mapping = impl.get_process_mapping()
  File "/usr/local/lib/python3.6/site-packages/pipenv/vendor/shellingham/posix/ps.py", line 18, in get_process_mapping
    'ps', '-ww', '-o', 'pid=', '-o', 'ppid=', '-o', 'args=',
  File "/usr/local/lib/python3.6/subprocess.py", line 336, in check_output
    **kwargs).stdout
  File "/usr/local/lib/python3.6/subprocess.py", line 418, in run
    output=stdout, stderr=stderr)
subprocess.CalledProcessError: Command '['ps', '-ww', '-o', 'pid=', '-o', 'ppid=', '-o', 'args=']' returned non-zero exit status 1.
*** Error code 1

Stop.

---

$ ssh user@freebsdhost "ps -o pid= || echo $?"
1

The above command is alternating 1 and 0 for some reason, I think because sometimes ps see a ps process, but I have been able to reproduce this quite reliably with Vagrant (run vagrant up then vagrant provision):

Vagrant.configure("2") do |config|
  config.vm.box = "generic/freebsd11"

  config.vm.provision :shell, privileged: false, inline: <<-EOF
    ps -ww -o pid=
    echo $?
  EOF
end

Thanks!

[uranusjr]

I guess we don’t really need to fail with a CalledProcessError here. If ps fails, we can’t detect shell anyway, so it’s better to swallow the process return code and just raise ShellDetectionFailure.

[uranusjr]

Workaround before Pipenv updates: Mount the procfs. Shellingham uses procfs to obtain process information if it is available, but on FreeBSD it is not enabled by default. How to do this.