Re: Bug#786902: O: ifupdown -- high level tools to configure network interfaces
On Wed, May 27, 2015 at 5:29 PM, Josh Triplett <josh@joshtriplett.org> wrote:
> On Wed, May 27, 2015 at 05:06:38PM -0700, Cameron Norman wrote:
>> On Wed, May 27, 2015 at 4:36 PM, Josh Triplett <josh@joshtriplett.org> wrote:
>> > Simon McVittie wrote:
>> >> One thing that an adopter could very usefully do with ifupdown would be
>> >> to coordinate with the systemd maintainers on moving net.agent
>> >> (Debian-specific udev glue to invoke ifupdown) from udev into ifupdown,
>> >> so that it does not need to be present at all on systems that rely on a
>> >> non-ifupdown tool like NM. That would also mean that the ifupdown
>> >> maintainer would be free to alter the precise details of how net.agent
>> >> and ifupdown interact, since they would now control both ends of the "API".
>> >
>> > I'd *love* to see that happen. I've seen discussions about that, and
>> > they always seemed to stall out.
>>
>> Then why not submit patches to ifupdown? It is not like it offers any
>> benefit to ifupdown users, so it is not really something the ifupdown
>> maintainer would be inclined to go out of his/her way to accomplish.
>> You can QA upload it really easily now too.
>
> Because I have no interest in developing ifupdown; I'd just like to see
> net.agent no longer run on systems *without* ifupdown.
Well then you just explained why the discussions seem to stall out:
people who will benefit (i.e. not ifupdown maintainers/users) have no
interest in touching ifupdown at all (at least it seems so).
--
Cameron Norman
Reply to: