monstrous failure of "yum update" on fedora 11 alpha

[Jerry Amundson]

On Thu, Mar 5, 2009 at 1:36 AM, seth vidal <skvidal at fedoraproject.org> wrote:
> On Thu, 2009-03-05 at 00:43 -0600, Jerry Amundson wrote:
>
>
>> Yes, we've established where we are now, but we need a path to be
>> better off in the the future. Heck even RHEL "up2date" is able to
>> update itself first, *and* be certain it will work afterwards.
>
> umm. No it isn't.
>
> up2date is in no better state than yum. It's in the same language using
> very similar calls.

So, if I steal the tires off a Porsche and put them on my Chevy Aveo,
my car will work just as well?

On Thu, Mar 5, 2009 at 1:58 AM, Jesse Keating <jkeating at redhat.com> wrote:
> On Thu, 2009-03-05 at 00:43 -0600, Jerry Amundson wrote:
>>
>> Yes, we've established where we are now, but we need a path to be
>> better off in the the future. Heck even RHEL "up2date" is able to
>> update itself first, *and* be certain it will work afterwards.
>
> Actually that fails more times than it works from what I hear.

My *experience* says otherwise. Yes, I realize there are well over 100
open up2date bugs. However, I have several RHEL systems on which I run
up2date monthly from cron, and often interactively using X, and I
can't recall the last time I had a problem. It would have been at
least a year ago. Whereas, with yum and Fedora, versions 10 and 11
especially, updates fail much more often.

Honestly, this is just worrying me that we throw the "yum update yum\*
rpm\*" answer to each problem, without an eye on the bigger prize (or
prizes) - stability, consistency, ease of use, and so on. The work
done on the graphical and system tray front-ends, fantastic that it
is, becomes moot in a flash when updates fail.

jerry