On Sun, 17.05.15 13:02, Martin Pitt (martin.p...@ubuntu.com) wrote: > Hey Lennart, > > Lennart Poettering [2015-05-14 18:09 +0200]: > > > As I mentioned before, simply ignoring /dev/root doesn't help in all > > > cases, and hardcoding it in the code is a bit ugly. > > > > It doesn't help in all cases? Which ones? Can you elaborate? > > It doesn't seem to help at all in e. g. LXC.
Sounds like borkage in LXC. Please ask LXC to follow these guidelines: https://wiki.freedesktop.org/www/Software/systemd/ContainerInterface/ If you follow these guidelines with your container software systemd will work fine. If you don't, then you are on your own. More specifically, they should follow the second item in the "Execution Environment" section: pre-mount /sys read-only in the container. Whether device management is available is detected by systemd by checking if /sys is writable. If it is writable it is assumed that the full device logic is available. This results in udev being started, .device units in systemd are made available and .mount units get dependencies on them. If LXC mounts /sys read-only however, then udev does not get started, no .device units are available, and no dependencies create on them. And all should work as intended. Lennart -- Lennart Poettering, Red Hat _______________________________________________ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel