On Wed, Jul 23, 2008 at 7:18 AM, Miklos Vajna <[EMAIL PROTECTED]> wrote:
> On Wed, Jul 23, 2008 at 07:05:43AM -0500, Dan McGee <[EMAIL PROTECTED]> wrote:
>> On Wed, Jul 23, 2008 at 3:01 AM, Miklos Vajna <[EMAIL PROTECTED]> wrote:
>> > On Tue, Jul 22, 2008 at 08:01:59PM -0500, Dan McGee <[EMAIL PROTECTED]>
>> > wrote:
>> >> # get setup correct
>> >> sudo rm
>> >> /var/lib/pacman/sync/{community,extra,pacman-git,testing,unstable}/.lastupdate
>> >
>> > Ouch, this is ugly. Why not using -Syy?
>>
>> Are you really critiquing a bisect script? I don't know why I'm giving
>> the time of day to this, but I'll bite.
>
> Should I next time just ask, without stating the reason I ask? ;-)
>
>> This script was designed to find the problem that occurred *only* on
>> -Syu and not -Su. Thus, a -Syy followed by and -Syu (which would
>> update zero databases) would not trigger the problem.
>
> I just thought -Syyu would work in that case, and wondered if there is a
> reason for doing so (read: learn from you as I suspected you know
> something that may be interesting).
Ahh, and I forgot to mention that the core .lastupdate was not
deleted. This was a key part of the equation if I remember right
because of interactions with packages also in testing.
-Dan
_______________________________________________
pacman-dev mailing list
[email protected]
http://archlinux.org/mailman/listinfo/pacman-dev
