[Battlemesh] WBMv8 -- a personal perspective

Henning Rogge hrogge at gmail.com
Mon Aug 10 15:15:51 UTC 2015


Hi,

some more thought by me on the "WiBed or Not WiBed" discussion.

WiBed has been very useful in earlier Battlemesh for quickly deploying
fixes to the firmware we had deployed. This time we didn't managed to
take advantage of this feature because we had trouble with WiBed.

If we can prevent this by having a WiBed version based on the latest
STABLE branch of OpenWRT next year, we can start testing much earlier
and then deploy some routing protocol upgrades based on the tests in
the middle of the week.

I would also like to present a configuration schema for the interfaces
of our routers... the proposed schema will have the following
attributes:

a) fixed prefix for each router
b) a sub-prefix which is distributed by a DHCP-server on EACH node to
the LAN port switch to attach measurement/debugging devices (e.g.
notebooks!)
c) no static routes necessary
d) looks the same from the testing point of view for the L3 protocols
AND for Batman-Adv.

It cost us quite a while to produce something similar last week, so if
we have a stable firmware AND a stable configuration, it should be
much easier to start early testing.

Henning

On Mon, Aug 10, 2015 at 4:57 PM, Benjamin Henrion <zoobab at gmail.com> wrote:
> On Mon, Aug 10, 2015 at 4:21 PM, Thijs van Veen <dicht at operamail.com> wrote:
>> Hi,
>>
>> Just to add in my thoughts.
>>
>> First of all, the shell scripts Henning and I wrote are very simple.
>> I thought we'd be doing much more difficult stuff, but it all turned out
>> pretty basic.
>> I hope people will learn from this and understand that it doesn't take a
>> genius to help out, just some focus.
>>
>> WiBed can be a very valuable tool to help us test things on a larger
>> scale and get more info from the nodes.
>> This is one of the reasons I offered Manos to help out with getting
>> WiBed up and running before the next Battlemesh.
>>
>> I'm pretty sure we can save a lot of issues just by grabbing the latest
>> stable versions and freeze that configuration for testing.
>> It's great to have this drive for testing the latest versions, but
>> please let us keep this seperate from the controlled testing
>> environment.
>> Any bugfixes after the frozen configuration should be acknowledged in
>> the discussion of the testresults.
>>
>> What I would like to see next year is basically three things:
>>
>> 1. A backup manual solution (possibly based on the scripts Henning and I
>> wrote);
>> 2. A WiBed setup with stable versions of OpenWRT and the protocols to be
>> tested;
>> 3. A WiBed setup with the bleeding edge versions of OpenWRT and the
>> protocols.
>>
>> All of this of course includes the protocol configurations, which may
>> depend on the test scenarios.
>> You can expect an email from me about a month before the event to gather
>> which scenarios we want to test and how we should configure the
>> protocols for these scenarios.
>
> Most of those problems would vaporize if the routers used at WBM would
> be part of a permanent testbed the rest of the year:
>
> http://battlemesh.org/BattleMeshV3.3/PermanentTestBed
>
> I am currently investigating ways to switch on/off those routers on
> USB mobile phone batteries via a side control channel hooked up to
> some arduino or similar. Serial port of the device could also be
> exposed through this control channel, which might be via wifi or
> 433mhz or another frequency, or via long cables.
>
> I also do not like indoor tests because of the amount of reflections
> involved compared to outdoor use.
>
> --
> 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."
> _______________________________________________
> Battlemesh mailing list
> Battlemesh at ml.ninux.org
> http://ml.ninux.org/mailman/listinfo/battlemesh


More information about the Battlemesh mailing list