juju show-status fails during teardown

[John A Meinel]

It is unclear if this is a bug in the test assumptions, or a bug in 'juju show-status', but we're having CI runs fail because of it:
http://qa.jujucharms.com/releases/6147/job/destroy-model-lxd/attempt/928
http://qa.jujucharms.com/releases/issue/5759a97d749a56227ed84d46

My first thought is that it is a problem with the test suite. Certainly if you do "juju show-status DYINGMODEL" and that model actually dies, we would expect status to fail, wouldn't we?

Maybe the model should still be there?
2018-01-11 12:33:00 INFO juju --show-log show-status -m destroy-model-lxd:test-tmp-env --format yaml
...
ERROR model "test-tmp-env" has been removed from the controller, run 'juju models' and switch to one of them.
...
raceback (most recent call last):
  File "/var/lib/jenkins/juju-ci-tools/assess_destroy_model.py", line 164, in <module>
    sys.exit(main())
  File "/var/lib/jenkins/juju-ci-tools/assess_destroy_model.py", line 159, in main
    assess_destroy_model(bs_manager.client)
  File "/usr/lib/python2.7/contextlib.py", line 24, in __exit__
    self.gen.next()
  File "/var/lib/jenkins/juju-ci-tools/deploy_stack.py", line 1056, in booted_context
    yield machines
  File "/usr/lib/python2.7/contextlib.py", line 24, in __exit__
    self.gen.next()
  File "/var/lib/jenkins/juju-ci-tools/deploy_stack.py", line 962, in runtime_context
    m_client.show_status()
  File "/var/lib/jenkins/juju-ci-tools/jujupy/client.py", line 1021, in show_status
    self.juju(self._show_status, ('--format', 'yaml'))
  File "/var/lib/jenkins/juju-ci-tools/jujupy/client.py", line 1150, in juju
    model, check, timeout, extra_env, suppress_err=suppress_err)
  File "/var/lib/jenkins/juju-ci-tools/jujupy/backend.py", line 197, in juju
    rval = call_func(args, stderr=stderr)
  File "/usr/lib/python2.7/subprocess.py", line 541, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '('juju', '--show-log', 'show-status', '-m', 'destroy-model-lxd:test-tmp-env', '--format', 'yaml')' returned non-zero exit status 1

That sure looks like were getting a non-zero return code, but because the model really isn't there.

Maybe its more the issue that "juju models" is listing models that are in a dying/dead state? There seem to be other bugs around that.

[John A Meinel (jameinel)]

I believe this was fixed by other fixes from Anastasia.