Thanks for the comments, Dan and Chris! I updated the MP accordingly.
SYSTEMD_OFFLINE will now take values '0' or '1' instead of 'true'. Originally I found out about this variable while reading the systemd code and I didn't look at the documentation. But, I agree that it is better to follow the doc.
I also liked the idea of `systemd_force_offline` defaulting to None so I implemented the change.
Lastly, I moved the systemd_force_offline variable to a more global location in `curtin.util.subp`. The logic is as follows now:
* if systemd_force_offline is True or False, then SYSTEMD_OFFLINE` is set accordingly in the environment (any previous value is overridden).
* if systemd_force_offline is None, then SYSTEMD_OFFLINE is set to '1' if and only if the two following conditions are met:
* We are executing the command in a chroot (i.e., target is not /)
* The SYSTEMD_OFFLINE variable is not present in the environment
Thanks for the comments, Dan and Chris! I updated the MP accordingly.
SYSTEMD_OFFLINE will now take values '0' or '1' instead of 'true'. Originally I found out about this variable while reading the systemd code and I didn't look at the documentation. But, I agree that it is better to follow the doc.
I also liked the idea of `systemd_ force_offline` defaulting to None so I implemented the change.
Lastly, I moved the systemd_ force_offline variable to a more global location in `curtin.util.subp`. The logic is as follows now:
* if systemd_ force_offline is True or False, then SYSTEMD_OFFLINE` is set accordingly in the environment (any previous value is overridden). force_offline is None, then SYSTEMD_OFFLINE is set to '1' if and only if the two following conditions are met:
* if systemd_
* We are executing the command in a chroot (i.e., target is not /)
* The SYSTEMD_OFFLINE variable is not present in the environment