Re: hotplug and ifup
On Wed, Feb 26, 2003 at 02:19:40AM +0900, Fumitoshi UKAI wrote:
> Anthony Towns wrote:
> > On Sun, Feb 23, 2003 at 11:55:52AM +1100, Herbert Xu wrote:
> > > Anthony Towns <email@example.com> wrote:
> > > > what should happen, and how should it happen? Should:
> > > hotplug should tell ifup that it is calling it, perhaps by using
> > > a special option. Then ifup can decide what the policy is.
Actually I think it would be good for the kernel to tell hotplug why
it's getting called ie -
"new driver (found existing hardware)"
"new device (existing driver found new device)"
Then hotplug can decide what to do according to the state of boot - or
pass that descision on to ifup and whatever other tools it uses to bring
Is this easy?
> > ifup can't be called before S39ifupdown has been run, which is the
> > problem here.
> Hmm, then should I fix at least hotplug should be activated
> after S39ifupdown? Maybe I should support blacklist (Bug#180521).
> > That aside, hotplug could quite easily call "ifup eth0=hotplug"; allowing
> > you to have "mapping hotplug" to handle any further details.
> Ok, I'll do this.
> I'm considering to modify hotplug to use run-parts mechanism (Bug#180017)
> which could also solve other bugs such as Bug#148590.
> Fumitoshi UKAI
> To UNSUBSCRIBE, email to firstname.lastname@example.org
> with a subject of "unsubscribe". Trouble? Contact email@example.com