Currently, the pull command returns 0 even when a failure occurred.
My usecase is having this running in a systemd timer job and every now and then it happens, that I shutdown my machine while there is a pull command running. This results in a lockfile being present which prevent future runs from succeeding.
However, since that case is caught in an exception there is the error output, but the job itself returns success (exit code 0).
This PR exits with code 1 as done in _try_load_config, as this was the simplest to implement. Maybe a nicer approach would be to raise a SystemExit instead but that will not directly put it to the critical log. Please let me know what you would prefer.
Currently, the
pull
command returns 0 even when a failure occurred.My usecase is having this running in a systemd timer job and every now and then it happens, that I shutdown my machine while there is a pull command running. This results in a lockfile being present which prevent future runs from succeeding.
However, since that case is caught in an exception there is the error output, but the job itself returns success (exit code 0).
This PR exits with code 1 as done in
_try_load_config
, as this was the simplest to implement. Maybe a nicer approach would be to raise aSystemExit
instead but that will not directly put it to the critical log. Please let me know what you would prefer.