Fedora 12: Make Dell machines boot again

[Jarod Wilson]

On Tue, Apr 6, 2010 at 3:43 PM, Adam Williamson <awilliam at redhat.com> wrote:
> On Mon, 2010-04-05 at 10:28 -0400, John W. Linville wrote:
>> On Sat, Apr 03, 2010 at 10:48:39PM +0200, Felix Schwarz wrote:
>>
>> > since the last stable kernel update in Fedora 12, a lot of Dell machines
>> > do not boot anymore due to a kernel oops... Of course bugzilla has a
>> > couple of bug reports (e.g. #578217, #579118, #578663, #578590).
>> >
>> > There is a known fix/workaround which is already present in CVS
>> > (kernel-2.6.32.10-94.fc12) and a koji build which fixes the issue for
>> > all users. However there's no update in bodhi yet.
>> >
>> > Due to the severity of the problem, I'd like to see a push to
>> > updates-testing as soon as possible.
>>
>> Thanks, Felix.  I have created the update.
>>
>> Since I created the problem for the b44 people in kernel -90, I
>> wanted to be sure to get some positive test reports before pushing
>> the -94 update.  I apologize for the extended delay over the weekend!
>
> From what I can tell, it looks like the -90 update got 'auto-pushed' by
> hitting +3 karma, despite the fact that two people had reported the
> regression in Bodhi. This is a classic case of 'works for me' positive
> feedback overriding 'there's a regression!' feedback, which is one of
> the issues we're trying to fix with the proposals to improve Bodhi.
>
> For now, might I suggest disabling auto-push for kernel updates?

Even more fun, the request was initially submitted and told to ignore
karma for auto-push, but the initial submission had an incorrect bug
number in it. Went in to fix that, and apparently didn't notice that
bodhi had helpfully decided to re-enable the auto-push check box when
all I wanted to do was fix a bug number. (in other words, "edit"
doesn't properly honor the current settings of the ticket).

-- 
Jarod Wilson
jarod at wilsonet.com