openssh: no pre-release sanity check? [Re: ssh-to-rawhide hangs

[Horst H. von Brand]

Clyde E. Kunkel <clydekunkel7734 at cox.net> wrote:
> On 09/11/2011 04:33 AM, Jim Meyering wrote:
> > darrell pfeifer wrote:
> >> Fails for me too, with the same error.
> >
> > Thanks for confirming that.
> >
> > I don't mean to be rude or inflammatory, but do have to wonder how such
> > a fundamentally-broken package was released -- even to rawhide.  In the
> > fedora openssh release process, isn't there some sort of sanity check
> > that would catch this before inflicting it on all rawhide users?

> Maybe there needs to be a classification for rawhide similar to the 
> karma system for updates-testing, but limited to just a set of packages 
> that should just always work (maybe openssh would be one).  For such 
> packages, there should be a test validation set that the maintainer runs 
> to automatically allow the package into rawhide.
> 
> I know in the past when such things have been discussed, some 
> maintainers chimed in that they do have test sets that they run on their 
> packages.

If that is so, that should count into whatever karma system is put in place.

> Doesn't have to be for all packages, just the ones that are critical or 
> necessary.  Yeah, defining those will be a challenge, but then what they 
> hey, makes life interesting.

The problem is that what is critical for me might be completely irrelevant
for you. I.e., TeXlive _has_ to work for me, if libreoffice fails to work
it can be a annoyance at worst; most users will wonder what TeXlive is and
scream bloody murder when libreoffice has hickups. On this (work) machine
the SSH server is a convenience (for logging in from the phone when away
from my desk for a fast check or such), on my "server" it is the only
convenient way to get in.

Perhaps the Critical Path packages
<https://fedoraproject.org/wiki/Critical_Path_Packages> need extra care
during their lifetime, not just leading up to release?  Perhaps OpenSSH
should be added (or placed in a separate "server critical" category, given
the above is mostly install/running/updating for "final users")?

BTW, I've seen several times that updating the server makes a running
instance non-responsive, but been too lazy to report it.
-- 
Dr. Horst H. von Brand                   User #22616 counter.li.org
Departamento de Informatica                    Fono: +56 32 2654431
Universidad Tecnica Federico Santa Maria             +56 32 2654239
Casilla 110-V, Valparaiso, Chile 2340000       Fax:  +56 32 2797513