[Commotion-dev] Questions about dashboard development

Seamus Tuohy s2e at opentechinstitute.org
Tue Oct 15 13:16:47 UTC 2013


Hey,

(see inline)

On 10/14/2013 05:18 PM, Ben West wrote:
> Howdy,
> 
> I'm writing after seeing this issue filed about collecting node tests
> results from remote nodes and presumably storing in a central location:
> https://github.com/opentechinstitute/commotion-dashboard-helper/issues/6
> 
> What are current thoughts about a centralized node database and/or
> dashboard for Commotion?  Any changes from this wiki page about the subject
> dated in June?
> https://code.commotionwireless.net/projects/commotion/wiki/BigBoard
> 

That page was the original brainstorming document for the dashboard we
kept. Development on the dashboard was then taken on by Nat, I don't
know where his documentation took place on the current dashboard. I do
know that we are using a variation of ffmap-d3
https://github.com/opentechinstitute/ffmap-d3 for our current dashboard
though.

> Offering comment on a high level, and after observing the ROBIN mesh
> project splinter into closed-source forks and dead ends, supporting
> interoperability with other mesh node dashboard and node database
> implementations is a big plus, even if only partial.
> 
> The Interop group have recently documented various schemas for a node
> databases here http://interop.wlan-si.net
> 
> By comparison, the CommonNodeDB repo from Aaron K (cited in the BigBoard
> wiki page) is now a couple years old:
> https://github.com/aaronkaplan/commonNodeDB
> 
> Additionally, WLAN Slovenia has their nodewatcher project, which appears to
> be at or nearing substantial maturity:
> https://dev.wlan-si.net/wiki/Nodewatcher
> https://github.com/wlanslovenija/nodewatcher
> 
> Maybe there is something interesting in Nodewatcher that could be a good
> fit for Commotion?

All the above is very true. I think that whatever dashboard/database
Commotion moves forward with in the future, the nodes themselves should
be the ones runnign tests upon the network and sending that information
back to the dashboard.

To that vein whatever dashboard we choose to move forward with should
allow us to query into the network for further information or test
suites from specific nodes without the adminstrator having to ssh into
the node to run them.

These are just my personal opinions of course. :)

s2e
> 
> 
> 
> _______________________________________________
> Commotion-dev mailing list
> Commotion-dev at lists.chambana.net
> https://lists.chambana.net/mailman/listinfo/commotion-dev
> 


More information about the Commotion-dev mailing list