:: Re: [DNG] Why Debian 8 Pinning is (…
Top Pagina
Delete this message
Reply to this message
Auteur: Rainer Weikusat
Datum:  
Aan: dng
Onderwerp: Re: [DNG] Why Debian 8 Pinning is (or isn't) pointless
Simon Hobson <linux@???> writes:
> Rick Moen <rick@???> wrote:
>
>> I have a better question: Is there something about empiricism that many
>> people on this mailing list cannot cope with?
>>
>> Back when I had newly joined this mailing list and all of these idle
>> allegations and rhetorical questions started being posted, I decided to
>> do that thing.... What's it called? Oh, right: 'Checking.'
>
> I too did some checking. From practical experience, one of the ClamAV
> packages (IIRC it's clamd) has a hard dependency on libsystemd0. Using
> dpkg --force-depends to install only that package without having
> libsystemd0 installed results in ... it failing at startup because it
> can't open the library.
>
> I opened a bug, which was very quickly and quite abusively closed as
> "won't fix", and was also told that "it doesn't work like that" when I
> asked if (especially as it was supposedly only one call they ever made
> on non-systemd systems) why they couldn't do "if exists libsystemd0
> then ..." - something which I now know is possible if the dev/packager
> cares about it.


It's possible but the point of libsystemd0 is reportedly putting this
logic (wrt systemd itself) into a library which can then be used by all
applications. Metaifying this to the next degree would ultimatively mean
create a libsystemd00 (ups --- still hard depedency), libsystemd000
(strangely, nothing changed), libsystemd0000 (still tied to systemd -
quel surprise) and so on.

A technical solution for this could be to create a way to interact with
'the service manager' in a way which doesn't require inserting calls to
functions provided by said service manager into applications, eg, by
executing a program (distant thump as systemd developer fainting upon
the supposition to fork and exec hits the floor) or by sending a message
in a documented format to some well-known address. An AF_UNIX datagram
socket would suggest itself for that.