[systemd-devel] Antw: Re: Antw: systemd.mount creating mount resource (What) for bind mounts

[Ulrich Windl]

>>> Lennart Poettering <lennart at poettering.net> schrieb am 13.01.2020 um 17:00
in
Nachricht <20200113160041.GB5677 at gardel-login>:
> On Mo, 13.01.20 12:03, Ulrich Windl (Ulrich.Windl at rz.uni‑regensburg.de)
wrote:
> 
>> > I see that a mount unit with `Options=bind` set creates the resource
>> > to be mounted, specified by `What`, in addition to the mount point,
>> > specified by `Where`, when they don't exist.
>>
>> Personally I think a mount operation should not create any missing 
> directory,
>> because a missing directory indicates some type of configuration problem 
> that
>> should be solved by hand.
> 
> We want to support cases where / is a tmpfs with only /usr mounted in
> from an OS image and where everything else is initialized and put
> together on boot.

How is that related? If you start from some initrd that initrd is the initial
root filesystem, obviously not empty. I think it's trivial to create the
mountpoints needed.


> 
> Similar, we want to support systems where / is reset via a "factory
> reset" concept (and thus empty), then combined with /usr and comes up
> properly initialized, with all mount points for /proc, /tmp, /sys and
> so on created.

I think there's more than one concept of "factory reset".

> 
> Lennart
> 
> ‑‑
> Lennart Poettering, Berlin