[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#727708: init system other points, and conclusion



On 01/04/2014 11:21 AM, Steven Chamberlain wrote:
Commenting as a porter, the decision on default init system might affect
me something like this:

If GNU/Linux defaults to Upstart, it's likely in porters' interest to
get that working as well as possible so we can keep consistency with
Linux arches.  I'm really grateful of Dimitri's work on this already.

But if GNU/Linux defaults to systemd, I'd say that's far too big, too
specialised around Linux, and likely to be a moving target to either
port it or maintain something compatible.

In that case, we may have to do the best we can with one of the other
init systems.  I'd wonder if it's still worth porting Upstart if few
systems would be using it, or packages having Upstart jobs.  I have good
feelings about OpenRC (which Gentoo already uses as an alternative
alongside systemd), or keeping plain sysvinit might even be still viable
for jessie only.

On Sat, 4 Jan 2014 14:09:57 +0800 Anthony Towns wrote:
I wonder if folks could clarify what status they expect secondary init
systems to have in Debian?
This aspect is most crucial to the ports.  At the very least, we'd need
to be able to get patches applied to fix startup issues on our init
system, even if the maintainer doesn't test or want to support this.  In
the worst case, we might have to give up on getting something like GNOME
working usefully without systemd, and thus not be able to ship it on
non-Linux ports.

Policy may need to explain whether hard systemd requirement is
permissible, if it should be expressed in package dependencies, or what
it should do otherwise (e.g. refuse to start, fail with error message,
fall back to something with reduced functionality).

If policy requires keeping functional sysvinit scripts around for
jessie, and/or (more controversially) can discourage the use of specific
non-portable functionality - which I think would be things like "expect
fork" or socket activation - I'm not necessarily saying this is a good
idea, but it would obviously work in our favour.

If non-Linux ports end up running and testing daemons on an alternate
init system *anyway*, I'd love for that work to benefit GNU/Linux users
who dislike the chosen default init system and want to use what we're
using instead.  And vice-versa, anyone running such a system and
finding/fixing startup issues, would likely be helping the ports.

[please keep me in Cc if responding directly to anything I said here]

Thanks,
Regards,
If Debian go's with systemd they need to use systemd 207 as its supported in RHEL 7 so we know it's going to be supported for around 10 years also why does Debian have systemd 204 in it's repos?? systemd 207 is way better


Reply to: