Roaming functionality with wpasupplicant (was: [pkg-wpa-devel] cleaning up our suggested modes of use)

[Kel Modderman]

On Thursday 03 August 2006 19:38, Felix Homann wrote:
> On Thursday 03 August 2006 09:59, Marc Haber wrote:
> > On Thu, Aug 03, 2006 at 11:24:49AM +1000, Kel Modderman wrote:
> > > I like Reinhard's idea of providing a framework for allowing such a
> > > plugin interface to provide "roaming logic". However, at this point in
> > > time, I do not consider logic provided outside of wpa_supplicant in any
> > > way superior, or even required. I simply find guessing the network,
> > > when it is already known, unnecessary.
>
> Kel's basically right here. Nevertheless, guessnet allows for a more
> unified way of choosing the right logical interface for several physical
> interfaces. For example, if a guessnet peer test (probing for IP/MAC
> addresses) is appropriate it doesn't matter if your using a wired or a
> wireless device.

Ok, maybe its time to investigate whether or not we can optionally use 
guessnet somehow. It could allow further flexibility to the roaming setup, 
but no idea how it can be done yet ;-)

>
> Since you can't use guessnet in combination with the wpa-roam solution you
> have to implement one mapping rule in wpa_supplicant.conf via id_str
> options and one for other devices in /e/n/i via the guessnet mechanism.
>
> Fortunately, "test ..." lines (necessary for guessnet) in a logical
> interface block don't disturb wpa-roam, so you can at least use one and the
> same logical interface with guessnet and wpa-roam managed devices.
>
> > It might be necessary when one roams between two dumbly-configured
> > networks, probably named "WLAN" or "NETG**R".
>
> You can work around this by specifying the BSSID in wpa_supplicant.conf.

Ah, good catch Felix. I must have misunderstood the question.

Say, while you attention is this way Felix, would you mind having a read of 
the doc thats being drafted?

http://svn.debian.org/wsvn/pkg-wpa/trunk/wpasupplicant/debian/README.modes?op=file&rev=0&sc=0

Any insight/opinion you would offer on it  would be valuable.

Thanks, Kel.