[Battlemesh] Network configuration and address plan for Battlemesh v9
nemesis at ninux.org
Tue Aug 11 18:17:38 CEST 2015
On 08/11/2015 06:01 PM, Dave Taht wrote:
> On Tue, Aug 11, 2015 at 6:00 PM, Dave Taht <dave.taht at gmail.com> wrote:
>> On Tue, Aug 11, 2015 at 5:44 PM, Henning Rogge <hrogge at gmail.com> wrote:
>>> On Tue, Aug 11, 2015 at 5:40 PM, Toke Høiland-Jørgensen <toke at toke.dk> wrote:
>>>> Henning Rogge <hrogge at gmail.com> writes:
>>>>> - ALL IPv4 addresses of the mesh are within a /16 prefix (so we don't
>>>>> need to push default routes to attached devices and can keep their
>>>>> Internet connection up)
>>>> Might I suggest adding IPv6 as well? Or an even more radical suggestion:
>>>> Do an IPv6-only test! :)
>>> Yes, I will add IPv6 now that we have resolved the IPv4 issues...
>>> I will post an update with IPv4 and IPv6 soon.
>> In laying out the original /21 design for ipv4, I left room for more
>> conventional prefix (/24) distribution, and more radios.
>> One of the things I would like to see tested more fully for the meshy
>> routing protocols is the more standard AP style multicast, which is
>> usually, today, distributed on a powersave (250ms) interval.
>> And it would be cool to be able to test devices with 3 or more radios,
>> which is increasingly common in the 802.11ac universe, and with things
>> that also connect to LTE and/or 802.14.
> And I also note that it might get easier to assign addresses, were
> devices to bootstrap and name themselves with the new hncp protocols.
Dave, I think we have to have honest and realistic expectations about
what we can accomplish.
I'd start simple and talk about more complex configurations and
scenarios when we'll realistically be able to replicate simple testbeds
with minimum effort.
More information about the Battlemesh