On Wed, Mar 17, 2010 at 2:44 PM, Daenyth Blank <daenyth+arch@gmail.com> wrote:
> On Wed, Mar 17, 2010 at 15:16, Aaron Griffin <aaronmgriffin@gmail.com> wrote:
>> I remember doing a redux of some of the backup array code way back
>> when, and I was fairly certain this was the case.
>>
>> Perhaps the tests of grub2 are hitting the condition in #2 here.
>>
>
> My understanding from reading the forum is that this is the case

In which case the behavior is normal and expected, yes? I mean, if I
understand the new grub2 correctly, there should be no config, but the
mkconfig command needs to be run.

On Wed, Mar 17, 2010 at 16:01, Aaron Griffin <aaronmgriffin@gmail.com> wrote:
> In which case the behavior is normal and expected, yes? I mean, if I
> understand the new grub2 correctly, there should be no config, but the
> mkconfig command needs to be run.
>

I think the issue was that the maintainer didn't realize removing it
from backup would remove it from user systems. I don't have the thread
at hand right now.

On Wed, Mar 17, 2010 at 3:03 PM, Daenyth Blank <daenyth+arch@gmail.com> wrote:
> On Wed, Mar 17, 2010 at 16:01, Aaron Griffin <aaronmgriffin@gmail.com> wrote:
>> In which case the behavior is normal and expected, yes? I mean, if I
>> understand the new grub2 correctly, there should be no config, but the
>> mkconfig command needs to be run.
>>
>
> I think the issue was that the maintainer didn't realize removing it
> from backup would remove it from user systems. I don't have the thread
> at hand right now.

Pacman should even handle that. The backup array is checked for both
the incoming package and the outgoing one, for stuff like this.

Now if it was never in the backup array to begin with, that's another story

On Wednesday, March 17, 2010, Aaron Griffin <aaronmgriffin@gmail.com> wrote:
> On Wed, Mar 17, 2010 at 3:03 PM, Daenyth Blank <daenyth+arch@gmail.com> wrote:
>> On Wed, Mar 17, 2010 at 16:01, Aaron Griffin <aaronmgriffin@gmail.com> wrote:
>>> In which case the behavior is normal and expected, yes? I mean, if I
>>> understand the new grub2 correctly, there should be no config, but the
>>> mkconfig command needs to be run.
>>>
>>
>> I think the issue was that the maintainer didn't realize removing it
>> from backup would remove it from user systems. I don't have the thread
>> at hand right now.
>
> Pacman should even handle that. The backup array is checked for both
> the incoming package and the outgoing one, for stuff like this.
>
> Now if it was never in the backup array to begin with, that's another story
>

The file was in the backup array in the old package and installed by
the pkgbuild. In the new package the file was still in the backup
array but not installed anymore. This either removed the file or moved
it to pacsave, I can't verify now which one of the two happened.
Obviously you are thrown into grub shell upon reboot. In the current
package i just install the file again, so people who still need ro
upgrade can boot normally. People who already did the upgrade should
intervene because the file was not under pacman control anymore.

Hope this makes it more clear what happened.

Ronald