No subject

time.
However we can perform the experiments one after the other. We just need to
adapt
the firmware to be able to switch the link-layer, and think about
not-concurrent experiments.

Cheers
--
./p4u


On Fri, Jan 25, 2013 at 7:41 PM, Thomas Pedersen <thomas at cozybit.com> wrote:

> Hi list,
>
> We'd like to represent o11s at WBMv6, but are a little concerned how
> the battlemesh will be done in practice. According to the wiki [1]
> there is a desire for all routing protocols to coexist on the same
> (physical) L2 link. As you probably know though, o11s implements its
> own routing protocol at L2 and cannot coexist with an IBSS interface
> on the same radio.
>
> One way to resolve this is let o11s be the "carrier" link for the
> different VLANs, but this would probably break the other routing
> protocols as they don't expect their L2 broadcast traffic to propagate
> throughout the entire mesh? Not to mention the mcast storm that would
> incur.
>
> We could set the o11s mesh TTL = 1 to ensure only single-hop
> propagation takes place, but of course this would have to be switched
> back when testing o11s and the tests could no longer be run in
> parallel.
>
> Another solution is to find a test platform that has 2 radios and
> dedicate one of them to o11s.
>
> Any ideas?
>
> --
> Thomas
>
> [1] http://battlemesh.org/BattleMeshV6/Firmware
> _______________________________________________
> Battlemesh mailing list
> Battlemesh at ml.ninux.org
> http://ml.ninux.org/mailman/listinfo/battlemesh
>

--f46d042f93507a135904d4236678
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Hello Thomas.<div>From my point of view o11s and the other protocols cannot=
 be tested at same time.=A0</div><div>However we can perform the experiment=
s one after the other. We just need to adapt</div><div>the firmware to be a=
ble to switch the link-layer, and think about not-concurrent experiments.</=
div>

<div><br></div><div>Cheers</div><div><div>--<br>./p4u<br></div>
<br><br><div class=3D"gmail_quote">On Fri, Jan 25, 2013 at 7:41 PM, Thomas =
Pedersen <span dir=3D"ltr">&lt;<a href=3D"mailto:thomas at cozybit.com" target=
=3D"_blank">thomas at cozybit.com</a>&gt;</span> wrote:<br><blockquote class=
=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padd=
ing-left:1ex">

Hi list,<br>
<br>
We&#39;d like to represent o11s at WBMv6, but are a little concerned how<br=
>
the battlemesh will be done in practice. According to the wiki [1]<br>
there is a desire for all routing protocols to coexist on the same<br>
(physical) L2 link. As you probably know though, o11s implements its<br>
own routing protocol at L2 and cannot coexist with an IBSS interface<br>
on the same radio.<br>
<br>
One way to resolve this is let o11s be the &quot;carrier&quot; link for the=
<br>
different VLANs, but this would probably break the other routing<br>
protocols as they don&#39;t expect their L2 broadcast traffic to propagate<=
br>
throughout the entire mesh? Not to mention the mcast storm that would<br>
incur.<br>
<br>
We could set the o11s mesh TTL =3D 1 to ensure only single-hop<br>
propagation takes place, but of course this would have to be switched<br>
back when testing o11s and the tests could no longer be run in<br>
parallel.<br>
<br>
Another solution is to find a test platform that has 2 radios and<br>
dedicate one of them to o11s.<br>
<br>
Any ideas?<br>
<span class=3D"HOEnZb"><font color=3D"#888888"><br>
--<br>
Thomas<br>
<br>
[1] <a href=3D"http://battlemesh.org/BattleMeshV6/Firmware" target=3D"_blan=
k">http://battlemesh.org/BattleMeshV6/Firmware</a><br>
_______________________________________________<br>
Battlemesh mailing list<br>
<a href=3D"mailto:Battlemesh at ml.ninux.org">Battlemesh at ml.ninux.org</a><br>
<a href=3D"http://ml.ninux.org/mailman/listinfo/battlemesh" target=3D"_blan=
k">http://ml.ninux.org/mailman/listinfo/battlemesh</a><br>
</font></span></blockquote></div><br></div>

--f46d042f93507a135904d4236678--