Re: upstart: please update to latest upstream version
On Thu, Feb 23, 2012 at 12:47:15AM +0000, Roger Leigh wrote:
> On Wed, Feb 22, 2012 at 02:58:50PM -0800, Russ Allbery wrote:
> > John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de> writes:
> >
> > > But I don't see the problem, Debian has the choice. We're not going to
> > > drop system V init anytime soon. Providing both systemd, upstart as well
> > > as classical system V init leaves the up to the user and allows to
> > > support non-Linux kernels.
> >
> > There's a serious drawback to supporting multiple init systems if one of
> > the goals is to stop writing init scripts. The only common denominator is
> > init scripts; upstart and systemd configuration files look entirely
> > different, and would have to be maintained separately if we support both
> > without using the init script compatibility support.
>
> Yes, there is absolutely a big cost to pay in supporting multiple
> init systems. Choice is good when there's a benefit, but we should
> not ignore the cost we pay.
>
> It would be good if we could generate init scripts from upstart
> and/or systemd service files, to permit migration to newer systems
> while still permitting the old system to be supported for the
> interim. It would IMO be more productive to port systemd and/or
> upstart to kFreeBSD/Hurd to make it possible to use the modern
> systems on all arches. The attitude of the systemd upstream is
> not encouraging here, however.
In sysv init scripts the daemon forks into the background. In upstrart
and systemd it doesn't have to (or shouldn't). (not) forking requires a
different command-line argument, normally. This leads to odd beasts such
as safe_mysqld.
--
Tzafrir Cohen | tzafrir@jabber.org | VIM is
http://tzafrir.org.il | | a Mutt's
tzafrir@cohens.org.il | | best
tzafrir@debian.org | | friend
Reply to: