[Discuss] SysVinit vs. systemd

[Bill Ricker]

On Thu, Sep 11, 2014 at 11:07 AM, Richard Pieri <richard.pieri at gmail.com> wrote:
> Ancient? Yes.
>
> Due for reform? Maybe. Depends on whether or not you consider lumping
> device management, system logging and superserver into the init system
> qualifies. I for one do not.

Yeah, I don't see how SysV init being slow / stiff / crufty / whatever
requires syslogd etc being replaced also; I didn't ask if you agreed
with that !

I'm assuming we're all agreed SYSV INIT needs at least major tweaks,
but there is disagreement on whether it should be fixed or replaced,
and how much of the rest of non-kernel system should be architected in
the same tranche of greenfield plowing if replaced.
     SystemD seems like an excessive solution to a larger problem
statement than we're agreed upon. But it's the only solution on offer
at the moment from the major Gnu/Linux distros.

>>   * Has Debian indicated what they intend to do for their BSD and Herd
>> kernel builds, since SystemD requires linux-specific kernel features?
>
> Debian will continue to use sysvinit with makefile concurrency for
> GNU/kFreeBSD. Debian/Hurd just recently switched from their homegrown
> init to sysvinit with the same mechanisms used in kFreeBSD.

Thanks for that info.

I wonder who will be the first to fork a sysvinit/syslog/... on an
unofficial Gnu/Linux variant of Debian, using the Debian BSD/Hurd
maintained versions of the legacy daemons, and if it will gain any
traction.

( From a security point of view, the Control Groups use in SystemD
*should* actually be a good thing, so I'm wait and see here.)

-- 
Bill Ricker
bill.n1vux at gmail.com
https://www.linkedin.com/in/n1vux