[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: problem with APT pinning



martin f krafft <madduck@debian.org> writes:

> also sprach Goswin von Brederlow <brederlo@informatik.uni-tuebingen.de> [2006.03.11.2241 +0100]:
>> Can you try pining the whole of http://people.ubuntu.com to 50 instead
>> of a per package pin?
>
> Done. Now it works.
>
> It seems like
>
>   Pin: release o=Jeff Bailey,l=Jeff Bailey
>
> doesn't work at all. In fact, I can't seem to pin on any o= and l=
> fields of the release data.

Right, that rings a bell.

>> Usualy the localy installed package has a pin of 100 so pins below 100
>> don't upgrade automatically. In your output I see that the local
>> source has pin 100, the ubuntu source pin 500 and the specific
>> packages have all pin 50 (thats the per package pin, right?).
>> 
>> So the latest ubuntu package wins out overall.
>
> It should not even apply. Default are pinned at 500, so any Ubuntu
> package will be 500. However, I specifically pin bzr to 50, and
> APT just ignores that.
>
> Bug report pending.

The problem is that it pins down the already installed package as
well. That should always be 100.

>> *** denotes the currently installed package:
>
> Where did you get this information? In my book, I describe *** as
> denoting the candidate and I am quite sure that I didn't "make that
> up" but empirically determined it; I did not go and look at the
> source though.

Just by looking at the output.

MfG
        Goswin



Reply to: