[Commotion-dev] Commotion Client Development

Dan Staples danstaples at opentechinstitute.org
Wed Jul 16 08:35:51 EDT 2014


Hey Joshua, just wanted to see if you've done any more experimenting
with the Commotion client? Unfortunately we haven't been able to do much
development on the client from our end, but if there are any small
things that would help you out with what you're doing, I'd love to help!

Dan

On 06/09/2014 04:11 PM, Dan Staples wrote:
> Responses inline:
> 
> On 06/07/2014 02:25 PM, Joshua Besneatte wrote:
>> Hi Paul,
>>
>> Thanks. This is exactly the project I was planning on starting myself,
>> luckily I found you guys and it's already going. The enthusiasm is easy
>> on my end, glad to see there are many others with my vision.
>>
>> I got a decent deb build working for Serval Project's serval-dna
>> incorporating a bunch of the scripts from OTI's serval-dna debian
>> folder. I can install/uninstall the serval-dna development fork,
>> including libs/includes/etc,  however, commotiond doesn't want to
>> compile against this version of serval-dna. Compilation of commotiond
>> errors out stating that the llseek64 system call is not available. So
>> far my only guesses are:
>>
>> 1. I am missing a dependency. (manpages-dev is supposed to provide
>> llseek64 but installing said package did not help)
>>
>> 2. commotiond needs to use OTI's version of serval-dna
>>
>> 3. I need to rebuild SP's serval-dna as package class "single" vs "indep"
>>
>> If anyone could shed some light on this, that would be great.
>>
>> If I build and install OTI's serval-dna I am able to compile commotiond
>> without problem.
> 
> The serval-dna plugin for commotiond is currently made to build against
> a specific version of libservald, which you can checkout in the
> 'commotion-wireless' branch of OTI's fork of serval-dna.
> 
> There is a _new_ version of the commotiond plugin that I'm working on
> that builds against a later version of libserval, which is in the
> 'refactoring' branch of commotiond. This branch should build against the
> 'serval-multi-key' branch of OTI's fork of serval-dna, though testing is
> still in progress.
> 
>>
>> As for the commotion-service-manager, does this still require
>> luci-commotion? Luci-commotion fails to install on requirement for the
>> deprecated serval-crypto.
> 
> Commotion-service-manager is also something I'm working on refactoring
> and upgrading, but I believe the master branch of CSM should still be
> compatible with the new versions of commotiond/serval-dna. The Debian
> packaging for CSM hasn't been updated to reflect the new dependencies
> however. If you're interested in doing that, that would be rad.
> Luci-commotion, like serval-crypto, is no longer a dependency. We should
> also add avahi-daemon to the Package dependencies, and
> libavahi-client-dev to the Source dependencies.
> 
>>
>> I have read most of the development docs and road-maps, but of course I
>> may have missed something. It seems the goal is to move to using SP's
>> serval-dna (for full serval compatibility) vs the OTI fork, is this
>> correct? Is OTI's serval-crypto deprecated because crypto is now handled
>> by SP's serval-dna?
> 
> We'd eventually like to use the Serval Project's debian package of
> serval-dna, but for now we have some incompatibilities we still depend
> on. Long term we'll probably drop the serval-dna plugin for commotiond
> and just use serval-dna itself. The only reason we don't do that now is
> because of size constraints on the router platform.
> 
>>
>> My nick is shamanon on IRC and I am in the #commotion channel if anyone
>> wants to strike up a conversation there.
> 
> Awesome. My nick is dismantl.
> 
>>
>> Thanks!
>>
>> On 06/07/2014 02:05 AM, Paul Gardner-Stephen wrote:
>>> Hi Joshua,
>>>
>>> Great work.  I recall that there may already be a debian package for
>>> serval-dna.  Andrew can probably comment on this.
>>>
>>> Paul.
>>>
>>>
>> _______________________________________________
>> Commotion-dev mailing list
>> Commotion-dev at lists.chambana.net
>> https://lists.chambana.net/mailman/listinfo/commotion-dev
> 
> Thanks for all the enthusiasm!
> 
> -Dan
> 

-- 
Dan Staples

Open Technology Institute
https://commotionwireless.net
OpenPGP key: http://disman.tl/pgp.asc
Fingerprint: 2480 095D 4B16 436F 35AB 7305 F670 74ED BD86 43A9


More information about the Commotion-dev mailing list