hrogge at gmail.com
Thu Dec 15 19:33:46 CET 2016
Just to say it,
if we can we should use the same hardware as last year, because two
Wifi interfaces one, dedicated ethernet plug and one 4-port switch we
can (VLAN) configure 3/1 is really good for experimentation.
On Thu, Dec 15, 2016 at 6:52 PM, Gui Iribarren <gui at altermundi.net> wrote:
> On 15/12/16 18:37, willi uebelherr wrote:
>> On 15/12/2016 07:24, Gui Iribarren wrote:
>>> I'm thinking out loud, that maybe with the librerouter we could set
>>> something up, when we have deployed them in the real networks. We'll
>>> have double-flash space, so we can put two firmwares (originally
>>> intended for safe sysupgrading, but could be used for "testbed" and
>>> "production" firmwares). Combining that with the hardware watchdog...
>>> Anyway, will come back to this idea in a few months when we have the
>> Dear Gui and all,
>> i searched for LibreRouter.
> totally unrelated
> totally unrelated as well
>> Open Router
>> LibreRouter project awarded FRIDA 2016 scale-up grant
> these two are correct
> as well as https://librerouter.org/
> sorry for the confusion
> re: benjamin regarding watchdog,
> it's a PIC that every 5 minutes decides to reboot the board or not. If
> it received a "ping" from userspace (simple bit up, bit down through a
> gpio), it won't reboot. Then, there's a very simple script running in
> userspace, sending that ping every 2 minutes for example. In case of a
> kernel panic or similar, the watchdog will not get pinged and will
> reboot the node.
> There's already the userspace software in available in openwrt, thanks
> to the implementation of such a watchdog in Open-Mesh routers.
>> It seems to me a big confusion. I agree and understand to develop the
>> router hardware itself. But from my experience in Latin America, also in
>> Argentina, never i have found a real strong process for hw-development.
>> And in Github Librerouter/Librekernel we find a large collection of
>> anything. But the task for a router is to route the packets through.
>> Therefore, the fisrst question is, based on what information we can
>> route. then, we can ask us, what hardware we need.
>> And the routers in our environment act mostly with radio
>> interconnection. But here we can abstract from the concrete transmission
>> technology and use a bidirectional connection to the transmitter device.
>> For cable connection we can use it directly.
>> This means, that in our space also the transmission technology stay in
>> the center. And i think, there we have the biggest challenges. because
>> if we look in the community networks reality, the lack of net-structure
>> is the biggest lack for community networks.
>> I don't understand this project.
>> many greetings, willi
>> Asuncion, Paraguay
>> Battlemesh mailing list
>> Battlemesh at ml.ninux.org
> Battlemesh mailing list
> Battlemesh at ml.ninux.org
More information about the Battlemesh