Bug#706522: Fwd: Bug#706522: ITP: pcs - Pacemaker/Corosync configuration system
Am 17.05.2013 um 09:42 schrieb Hadret <hadret@gmail.com>:
> On Fri, May 17, 2013 at 09:21:09AM +0200, Martin Gerhard Loschwitz wrote:
>> Filip,
>> Am 17.05.2013 um 09:15 schrieb Filip Chabik <hadret@gmail.com>:
>> in fact, I had already uploaded packages to experimental (I just got a REJECT on
>> them yesterday for some minor things though).
>
> This is awesome news (that you already did it, not that it was
> rejected). I have fingers crossed for this to succeed next time you
> try.
>
Ya. It really wasn't anything big, but pcsd scares the shit out of me a little bit, esp. because
it tries to download something during the build process, which looks rather scary. Oh well.
>> Have you included pcsd? That looked highly broken the last time I looked at it.
>
> As far as I saw (I removed packages and my built once you told me, that
> you already done it), pcsd was included in the packages -- thing is, I
> had no place to test it out :(
>
Ouch :\
>> I have actually packaged the ruby gems required for pcsd already. It's just that
>> pcsd looks like something that will blow up every second, so I did not include it
>> in the package I originally uploaded to experimental.
>
> Is it the approach suggest for package maintainers? I'm asking cause I
> wonder whether there's no conflict in installing gems via gem command
> and APT?
>
Well, the general assumption is: If a package needs a ruby gem, that gem would need
to be packaged as a dependency. Installing ruby-gems via postinst is something I
would rather not want to do; I don't know if it is even allowed as per policy.
The Ruby packages I produced are here:
http://people.debian.org/~madkiss/ruby/
I'd be happy if somebody could discuss with the Ruby maintainers how to carry on with
this; e.g. would there be any interest at their end to maintain these?
> Filip
Best regards
Martin
Reply to: