[Battlemesh] About a testbed - what is the status

[Benjamin Henrion]

On Tue, Jul 28, 2015 at 9:03 PM, Mitar <mitar at tnode.com> wrote:
> Hi!
>
> Could we assure this things work before the event? Not that we spend
> debugging the issues during?
>
> So could you compile images for all the devices we are having at the
> event so that people can try them in advance?
>
> Also, which documents are there now for what would be the topology of
> the testbed? What I worry is that we should not be generating the
> traffic with same nodes we are doing routing with. So is this separated?
> Some nodes are doing only routing, other have a independent WiFi channel
> to communicate control plane over, and then we have dedicated edge
> machines which generate test traffic.

Unfortunately I won't be able to come this year, but I wanted to
propose a simple test to discard all those routing protocols that do
not understand multiple channels, and rely on simple packet loss tests
to decide which route to take.

As I explained it at WBMv3, staying on the same channel should be
mostly then switching of channel.

So a simple D test like this:

A
|
B-D
|  |
C-E

where:

* all the 5 nodes are 1 radio, 1 ethernet)
* B-D and C-E are ethernet cabled
* A-B-C is channel1 2.4Ghz
* D-E is channel 11 2.4Ghz

Additionally, there should be 2 ethernet traffic generators (iperf or
other) connected to A and C so that routers are only routing.

Intelligent routing protocols should do A-B-D-E-C and not A-B-C.

Babel with the -z feature should behave like this, it is in Babel
since post-WBMv3, but it was never really tested.

If someone has the time to implement this testbed, I would be
interested by the results.

Now I don't know if it can be modified to fit dual-band APs, or if B-D
can be merged into our router.

Have fun next week,

--
Benjamin Henrion <bhenrion at ffii.org>
FFII Brussels - +32-484-566109 - +32-2-4148403
"In July 2005, after several failed attempts to legalise software
patents in Europe, the patent establishment changed its strategy.
Instead of explicitly seeking to sanction the patentability of
software, they are now seeking to create a central European patent
court, which would establish and enforce patentability rules in their
favor, without any possibility of correction by competing courts or
democratically elected legislators."