[arch-dev-public] [signoff] libgpg-error 1.6-1 + libgcrypt-1.4.0-1

[Jan de Groot]

On Tue, 2008-01-08 at 09:40 -0600, Aaron Griffin wrote:
> On Jan 8, 2008 9:22 AM, Dan McGee <dpmcgee at gmail.com> wrote:
> > <http://bugs.archlinux.org/task/9114>
> >
> > So...we have some issues here, and this got moved to core. I wanted to
> > go back and see who signed off to see who tested, and am quite
> > surprised to see not a single signoff for i686. Why on earth did this
> > get moved out of testing?
> >
> > Usually I don't like to point fingers at a single person, but we had a
> > clear breakdown of policy here, and its hard to spread the blame.
> 
> Thanks for pointing this out Dan. It is true. We didn't have any
> signoffs here for i686. I brought up the topic of testing, but do not
> actually use any of the encryption and pgp type stuff, so wouldn't
> even know where to begin.
> 
> This little testing policy was put in place for a reason. Not to make
> people's lives harder, but, in fact to make them easier. Less errors
> are a good thing. And the first time we let our policy collapse, we
> have (according to that bug report) at least 3 packages crashing due
> to this update that was pushed to core inappropriately.
> 
> Andy, next time please be more careful. Now we have to mop up a mess
> created by being too hasty.
> 
> So, do we rollback the libgcrypt in core, or do we wait? Jan, you
> appear to have done some investigation - could you fill us in?

libgcrypt 1.4.0-1.1-i686.pkg.tar.gz is in core now. I disabled VIA
padlock support and one of the reporters confirmed the bug as fixed with
this version.