> I originally created this branch because I was tired of guessing what was
> wrong when juju-deployer error message was only "returned with non-zero
> status". I've seen the code in tests/run.py, I think the "way to do this" you
> mention is needs_bootstrap in tests/run.py, which is a bit more sophisticated
> than the simple check I'm performing right now. It seemed a bit too much to
> accomplish what I wanted here but maybe we should just go ahead and make the
> same validations in both places, as we need similar setups to be able to
> deploy and test.
That's a fair point. Its hard to argue with a patch that still makes things better.
> I originally created this branch because I was tired of guessing what was
> wrong when juju-deployer error message was only "returned with non-zero
> status". I've seen the code in tests/run.py, I think the "way to do this" you
> mention is needs_bootstrap in tests/run.py, which is a bit more sophisticated
> than the simple check I'm performing right now. It seemed a bit too much to
> accomplish what I wanted here but maybe we should just go ahead and make the
> same validations in both places, as we need similar setups to be able to
> deploy and test.
That's a fair point. Its hard to argue with a patch that still makes things better.