On Mon, Apr 03, 2000 at 08:49:17PM +1000, Craig Sanders wrote:
> On Mon, Apr 03, 2000 at 09:46:13AM +0100, Julian Gilbey wrote:
> > On Mon, Apr 03, 2000 at 01:22:12PM +1000, Craig Sanders wrote:
> > > debian 'unstable' is perfectly usable for production servers, using
> > > it for such does not require any more caution about upgrades than
> > > using debian 'stable' or debian 'frozen'.
> >
> > Like during the Perl transition period, or when a recent libstdc++
> > broke apt, or when su stopped being able to su, or when ....
> >
> > Need I continue?
>
> i repeat: "[using unstable] does not require any more caution about
> upgrades than [using stable]"
>
> upgrading to whatever the latest stable releases is requires just as much
> caution/paranoia as upgrading to whatever is in the latest unstable. anyone
> who trusts the latest debian stable release on their critical/production
> servers without testing it on other machines first deserves whatever they
> get.
>
> if you have a clue and you are cautious then both stable and unstable are
> safe. if you don't have a clue or you are not cautious, then neither are.
A very good point. I've had breaks before now with permissions on /dev/ptmx,
for example ("stable"? It was stormix I started from ;) going to 'frozen' -
they can happen *between versions* as well as "is using unstable".
If we weren't up to fixing the odd occasional buglet, we wouldn't be
dist-upgrading against either frozen or unstable every day religiously ;]
It's the ones where the pre-configure script calls a binary giving "exec
format not recognised" that crack me up ;8)
~Tim
--
| Geek Code: GCS dpu s-:+ a-- C++++ UBLUAVHSC++++ P+++ L++ E--- W+++(--) N++
| w--- O- M-- V-- PS PGP++ t--- X+(-) b D+ G e++(*) h++(*) r--- y-
| The sun is melting over the hills, | http://piglet.is.dreaming.org/
| All our roads are waiting / To be revealed | piglet@glutinous.custard.org