Gabor Gombas wrote:
> On Mon, Oct 09, 2006 at 01:44:00AM +0200, Michael Biebl wrote:
>
>> We don't really need the ability to install *multiple* init systems in
>> parallel imho
>
> Yes we do, for the same reason we allow multiple kernel images to be
> installed simultaneously: if the new one doe
* Matthias Julius ([EMAIL PROTECTED]) wrote:
> Eric Dorland <[EMAIL PROTECTED]> writes:
>
> > * Matthias Julius ([EMAIL PROTECTED]) wrote:
> >> Eric Dorland <[EMAIL PROTECTED]> writes:
> >>
> >> >> - If you set up the alternatives in preinst, then there is a time when
> >> >> the symlink exists
Eric Dorland writes ("Re: anticipating the upstart migration"):
> Shouldn't it be possible to move the alternatives around in an atomic
> fashion? ln -sf bar foo.tmp ; mv foo.tmp foo . Or am I missing
> something?
This is in theory possible, I suppose. You wo
On Oct 11, Michael Biebl <[EMAIL PROTECTED]> wrote:
> Having the "init" binary installed as /sbin/upstart and only diverting
> the not so critical binaries seems to be the safest option indeed.
I had to use many diverions in the module-init-tools package because
there was no other acceptable solut
Eric Dorland <[EMAIL PROTECTED]> writes:
> * Matthias Julius ([EMAIL PROTECTED]) wrote:
>> Eric Dorland <[EMAIL PROTECTED]> writes:
>>
>> >> - If you set up the alternatives in preinst, then there is a time when
>> >> the symlink exists but the pointed binary hasn't been unpacked yet ->
>> >>
* Matthias Julius ([EMAIL PROTECTED]) wrote:
> Eric Dorland <[EMAIL PROTECTED]> writes:
>
> >> - If you set up the alternatives in preinst, then there is a time when
> >> the symlink exists but the pointed binary hasn't been unpacked yet ->
> >> unbootable system.
> >> - If you set up the alte
Henrique de Moraes Holschuh wrote:
> On Mon, 09 Oct 2006, Gabor Gombas wrote:
>> On Sun, Oct 08, 2006 at 10:53:39PM -0300, Henrique de Moraes Holschuh wrote:
>>> No wrappers for the single most critical binary in a Unix system after the
>>> libc. Sorry.
>> Right. How about upstart not providing a
Eric Dorland <[EMAIL PROTECTED]> writes:
>> - If you set up the alternatives in preinst, then there is a time when
>> the symlink exists but the pointed binary hasn't been unpacked yet ->
>> unbootable system.
>> - If you set up the alternatives in postinst, there is a time when there
>> is
* Gabor Gombas ([EMAIL PROTECTED]) wrote:
> On Tue, Oct 10, 2006 at 05:25:57PM -0400, Eric Dorland wrote:
>
> > Shouldn't it be possible to move the alternatives around in an atomic
> > fashion? ln -sf bar foo.tmp ; mv foo.tmp foo . Or am I missing
> > something?
>
> - If you set up the alternat
On Tue, Oct 10, 2006 at 05:25:57PM -0400, Eric Dorland wrote:
> Shouldn't it be possible to move the alternatives around in an atomic
> fashion? ln -sf bar foo.tmp ; mv foo.tmp foo . Or am I missing
> something?
- If you set up the alternatives in preinst, then there is a time when
the symlink
* Ian Jackson ([EMAIL PROTECTED]) wrote:
> Gustavo Noronha Silva writes ("Re: anticipating the upstart migration"):
> > The alternatives system is quite mature; it suffered from leaving
> > dangling alternatives, but that was ages ago...
>
> The alternatives system
Gustavo Noronha Silva writes ("Re: anticipating the upstart migration"):
> The alternatives system is quite mature; it suffered from leaving
> dangling alternatives, but that was ages ago...
The alternatives system must not be used for any of the essential
files of an essential p
On Tue, 10 Oct 2006, Reinhard Tartler wrote:
> /var/{run,lock} could be mounted as tmpfs in early userspace. Other
> distributions are already doing this, and a few weeks ago, there was
> discussion about doing this in debian as well.
For various reasons, Debian will go with /lib/init/rw as the ea
Alex Pennace <[EMAIL PROTECTED]> writes:
> On Sun, Oct 08, 2006 at 10:16:51PM +0200, Hendrik Sattler wrote:
>> I propose another solution. Introduce init-common with wrappers:
>> * /sbin/init is a binary that
>> - reads a configuration file with the init system name and
>> - creates a file /v
On Sun, Oct 08, 2006 at 10:16:51PM +0200, Hendrik Sattler wrote:
> I propose another solution. Introduce init-common with wrappers:
> * /sbin/init is a binary that
> - reads a configuration file with the init system name and
> - creates a file /var/run/inittype (or whereever is can be stored a
On Mon, 09 Oct 2006, Gabor Gombas wrote:
> On Sun, Oct 08, 2006 at 10:53:39PM -0300, Henrique de Moraes Holschuh wrote:
> > No wrappers for the single most critical binary in a Unix system after the
> > libc. Sorry.
>
> Right. How about upstart not providing a /sbin/init binary at all, but
> inst
On Mon, Oct 09, 2006 at 01:44:00AM +0200, Michael Biebl wrote:
> We don't really need the ability to install *multiple* init systems in
> parallel imho
Yes we do, for the same reason we allow multiple kernel images to be
installed simultaneously: if the new one does not work, there should be
a wa
On Sun, Oct 08, 2006 at 10:53:39PM -0300, Henrique de Moraes Holschuh wrote:
> No wrappers for the single most critical binary in a Unix system after the
> libc. Sorry.
Right. How about upstart not providing a /sbin/init binary at all, but
instead using an "init=/sbin/upstart" boot parameter? Th
On Sun, 08 Oct 2006, Hendrik Sattler wrote:
> I propose another solution. Introduce init-common with wrappers:
No wrappers for the single most critical binary in a Unix system after the
libc. Sorry.
--
"One disk to rule them all, One disk to find them. One disk to bring
them all and in the
Gustavo Noronha Silva wrote:
> Em Sun, 8 Oct 2006 18:56:39 +0200
> Hendrik Sattler <[EMAIL PROTECTED]> escreveu:
>
>>> The various commands you said could be provided as slaves to
>>> the /sbin/init alternative, which will make them be switched
>>> 'atomically'.
>> But you need the shutdown comman
On Oct 08, Hendrik Sattler <[EMAIL PROTECTED]> wrote:
> I propose another solution. Introduce init-common with wrappers:
I propose that people here will finally learn that complexity is bad.
Your idea makes me want to cry.
--
ciao,
Marco
signature.asc
Description: Digital signature
On Mon, Oct 09, 2006 at 12:45:17AM +0200, Gabor Gombas wrote:
> PostgreSQL did a similar trick because upgrading to a new major version
> needs the binaries of the _previously_ installed version to be able to
> upgrade the database.
And this was extremely error-prone and caused all sorts of failur
On Sun, Oct 08, 2006 at 11:48:44PM +1000, Anthony Towns wrote:
> What happens if you lose power between deinstalling sysvinit and unpacking
> upstart, eg?
Always leave the old version of init available as /sbin/init.old. That
way you can get a working system by booting with "init=/sbin/init.old"
Am Sonntag 08 Oktober 2006 21:09 schrieb Gustavo Noronha Silva:
> Em Sun, 8 Oct 2006 18:56:39 +0200
>
> Hendrik Sattler <[EMAIL PROTECTED]> escreveu:
> > > The various commands you said could be provided as slaves to
> > > the /sbin/init alternative, which will make them be switched
> > > 'atomical
Em Sun, 8 Oct 2006 18:56:39 +0200
Hendrik Sattler <[EMAIL PROTECTED]> escreveu:
> > The various commands you said could be provided as slaves to
> > the /sbin/init alternative, which will make them be switched
> > 'atomically'.
>
> But you need the shutdown command for the running init, not for t
Em Sun, 08 Oct 2006 18:57:50 +0200
Michael Biebl <[EMAIL PROTECTED]> escreveu:
> > The various commands you said could be provided as slaves to
> > the /sbin/init alternative, which will make them be switched
> > 'atomically'.
>
> Ok, but it would still need changes to the sysvinit package, which
Am Sonntag 08 Oktober 2006 18:40 schrieb Gustavo Noronha Silva:
> Em Sun, 08 Oct 2006 18:25:57 +0200
>
> Michael Biebl <[EMAIL PROTECTED]> escreveu:
> > > "sysvinit only" -> "sysvinit + upstart" (using
> > > alternatives/diversions)
> >
> > No offense, but I wouldn't trust the alternatives syst
Gustavo Noronha Silva wrote:
> Em Sun, 08 Oct 2006 18:25:57 +0200
> Michael Biebl <[EMAIL PROTECTED]> escreveu:
>>> "sysvinit only" -> "sysvinit + upstart" (using
>>> alternatives/diversions)
>> No offense, but I wouldn't trust the alternatives system for something
>> sensible as /sbin/init. Pl
Em Sun, 08 Oct 2006 18:25:57 +0200
Michael Biebl <[EMAIL PROTECTED]> escreveu:
> > "sysvinit only" -> "sysvinit + upstart" (using
> > alternatives/diversions)
>
> No offense, but I wouldn't trust the alternatives system for something
> sensible as /sbin/init. Please also remember that it's not
Anthony Towns wrote:
> On Fri, Oct 06, 2006 at 10:43:00PM +0200, martin f krafft wrote:
>> In order to enable this change, we're facing the "To continue type
>> in the phrase 'Yes, do as I say!'" problem because sysvinit is
>> marked essential.
>
> That's only a problem if you're going from "sysvi
On Fri, Oct 06, 2006 at 10:43:00PM +0200, martin f krafft wrote:
> In order to enable this change, we're facing the "To continue type
> in the phrase 'Yes, do as I say!'" problem because sysvinit is
> marked essential.
That's only a problem if you're going from "sysvinit only" to "upstart
only" as
also sprach Steve Langasek <[EMAIL PROTECTED]> [2006.10.08.0846 +0200]:
> As enthusiastic as I am about being able to replace sysvinit with
> something better, this is the wrong time in the release cycle to
> be making dependency changes to Essential, or even to the base
> system in general. I cer
On Fri, Oct 06, 2006 at 10:43:00PM +0200, martin f krafft wrote:
> upstart is looking interesting and it might just as well replace
> sysvinit for etch+1. Or at least be an alternative.
> In order to enable this change, we're facing the "To continue type
> in the phrase 'Yes, do as I say!'" proble
upstart is looking interesting and it might just as well replace
sysvinit for etch+1. Or at least be an alternative.
In order to enable this change, we're facing the "To continue type
in the phrase 'Yes, do as I say!'" problem because sysvinit is
marked essential.
Jeroen said this has been discus
34 matches
Mail list logo
