Network association algorithm

Miguel Álvarez miguel at
Wed May 2 14:20:39 EDT 2007

Hi Javier, (and everyone else)

The proposed procedure doesn't detail how the mpp detection should work
(which is to mean that currently it is supposed to use the customized udp
package). It only specifies the order of each step, and which possible
connectivity scenarios may arise.

In that respect, I would suggest to change the order of the dhcp and mpp
queries, so as to finish in less time: (I am taking the time brackets
written on the whiteboard, so it may be inaccurate)

Every DHCP attempt takes around 30s, while the assignation of a IPv4LL
address + MPP discovery only take around 11s. It would be possible to
incorporate some kind of flag to the reply message, to specify if the mpp
contacted is a school server.*
This would reduce the elapse time per failed attempt by 19s, without making
any changes to the rest of the diagramm.

Then there is the issue of changing from the current mpp discovery protocol
to a 'more standard' way of doing it; but appart from 'standardisation' (and
showing anyone with dns-sd capabilities the presence of the server), I
cannot say what the advantages would be



* If there is a school server present, we can suppose that no other node on
the mesh has mpp status

On 5/2/07, Javier Cardona <javier at> wrote:
> Hi John, Dan,
> > 1. The XO finds an established MPP (w. DHCP), and uses it for
> > internet access.
> > 2. The XO finds an AP, and becomes an ad-hoc MPP (no DHCP)
> > 3. The XO finds an ad-hoc MPP (no DHCP), and uses it for internet access
> > 4. The XO finds no AP nor MPP, and becomes a mesh node on channel 1
> > w. no internet access
> Some questions...
> 1. Is this algorithm using the MPP anycast address to discover the best
> MPP?
> 2. Is DHCP replacing the non-standard MPP requests used by mpp-utils?
> I was contacted a while ago by Miguel Alvarez (miguel at who
> is looking for a more "standards compliant" mechanism to configure
> MPs.  He suggested to replace the custom mppreqs with DNS-SD.  Has he
> been part of these discussions?
> Cheers,
> Javier
> --
> Javier Cardona
> cozybit Inc.
> p 415 974 6770
> f 415 974 6771
> c 415 630 0627
> e javier at
> _______________________________________________
> Devel mailing list
> Devel at

violence is the last refuge of the incompetent
--isaac asimov
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the Devel mailing list