Re: Running systemd with PID != 1, coexisting with other inits
Hi,
Josh Triplett:
> Well, *that's* useful; thanks! I previously had the impression that
> systemd did not support this at all.
>
Also see systemd-nspawn(1), a chroot-on-steroids which can "boot" an
entire subdirectory (like the result of 'debootstrap') without messing
around with /usr/sbin/policy-rc.d and/or endangering the rest of your
system (broken sys5rc files with "killall" in them, network reconfigu-
ration, etc.).
> This seems like a sensible, sustainable, long-term solution for
> supporting multiple init systems as PID 1, while still allowing services
> to make use of systemd-specific functionality. (Much like services
> today could depend on runit.) Thoughts?
>
You'd need to teach this sub-systemd to not run units for which a sys5rc
file exists, if sys5rc happens to be active. Other than that (and some
research about the cgroups issue) I suspect that the main problem would
be social rather than technical, for reasons I suppose I don't need to
elaborate.
--
-- Matthias Urlichs
Reply to: