Bug#727708: systemd jessie -> jessie+1 upgrade problems
On Wed, Dec 18, 2013 at 02:53:39PM -0800, Russ Allbery wrote:
> Adrian Bunk <firstname.lastname@example.org> writes:
> >  Personally, I am sceptical whether it is a good idea to switch to a
> > different init system for jessie. But I am not on a desperate rant
> > against systemd, and if something I bring up can be addressed that
> > is positive for me.
> Just to give fair warning: right now, based on what I know today, there is
> basically zero chance that I personally will vote retaining sysvinit for
> jessie above further discussion. So if you want to convince at least this
> one member of the technical committee that this is a viable option, you
> have quite a bit of work to do.
Where would "further discussion in 1-2 years" rank for you?
What I suggested earlier in this discussion was not that you vote for
keeping sysvinit forever, but:
* jessie will continue to use sysvinit, and the TC will re-evaluate
the situation after the release of jessie
As time passes, the kernel<->systemd interface will become more
stable since new features systemd wants like kdbus will already
be in the kernel, reducing potential upgrade problems.
It will also become more clear how exactly systemd is evolving and what
exactly the consequences are for Debian in areas where Debian differs
from other distributions.
And in my opinion the worst case would be that Debian switches init
systems in jessie and then again in jessie+1.
OpenRC looks too WIP and with an unclear future for me for switching
to it for jessie.
Upstart is mature, but I would be cautious since Canonical might decide
at some point in the future that systemd is better and abandon upstart.
I am not seeing a big probability of that happening, but it is a risk.
If you are convinced that any of systemd/OpenRC/upstart is the best
option for Debian then vote for it, but no matter where you vote
"further discussion" it is unlikely that there will be significant
new arguments in the immediate future - and it would therefore be
better to wait 1-2 years until the further discussion happens.
> Right now, it appears to me that the feature set is wholly inadequate,
> further substantial development is highly unlikely, the configuration
> syntax is familiar but awful, and we are already seeing software in the
> archive that requires capabilities that it's just not able to support.
> To support continuing to use sysvinit, I think I would need to see a
> credible plan for significant upstream development and support of the
> software, including, at a minimum, how the features that are required by
> our desktop environments would be handled, how device management and
> dependencies will be managed given the event-driven kernel, and how proper
> daemon management at least at the level common to upstart, systemd, and
> OpenRC will be added.
When staying with sysvinit is only the stopgap solution for jessie, then
significant upstream development would not even bring any advantages.
> Russ Allbery (email@example.com) <http://www.eyrie.org/~eagle/>
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed