Tiny bits of information mean so much...
"That should work. In the first boot it takes a longer time to respond to a DHCP request and to a ssh request. To be sure reboot the NS."
Yeah, it took a while, and wanted rebooting. Now I can SSH in.
In the first boot, after the device is flashed entirely, or just the rootfs or the rootfs_data is erased, mini_fo does its magic and it makes the boot process slower. Then, other boots are faster.
As I'm wandering through it, I am very impressed - it seems to work very well so far!
Thanks.
I'm very curious about why each node has a separate SSID. I was drawn to Nightwing because I like the general function of Open-Mesh (RoBIn) but, since I want to put in outdoor mesh networks, I want "beefier" radios. Open-mesh, of course, gives each node the same 2 SSIDs, gleaned from the "dashboard" config. This way one can have a degree of "mobility" between nodes - when one passes out of range of one node, the client radio will seek the same SSID and reconnect.
This was introduced because we got some reports that said that the Windows OS had problems when it detected 2 AP with an unique SSID (it stucks with the first connected AP and didn't want to connect to the other AP wich was what it should do).
This is been discussed in the development team.
Is there a way to easily tie Nightwing to a "gateway" portal (Orangemesh, open-mesh, etc.)?
Right now we don't use what is call a Dashboard. We use WiFiDog Auth Server to do some tasks that the other software do. We also have content filtering implemented with OpenDNS (this last one and other features are in the next release).
Basically, we are adding features to Nightwing that we are going to use. But leaving the possibility to others to add or subtract what they want to suit their needs.
Interestingly, Open-mesh has moved entirely to OLSR instead of B.A.T.M.A.N. - any thoughts about that for Nightwing?
When we moved to B.A.T.M.A.N., after testing RoofNet, Meraki and a lot of other related projects, we also took a look at OLSR but discard it. It was discarded for technical reasons such as rooting loops problems, protocol overhead, etc; and also for philosophical reasons such as license used, community concept, etc.
Regarding open-mesh.com moving to OLSR and the reasons that I've read, I can only say that I don't share them. The tests we've done and other reports that we had showed us different results.
Thanks, and I'll stay in touch and share my experiences here.
Cool.
To get more info related to the project, you can subscribe to the mailing lists:
http://www.lugro.org.ar/mailman/listinfo/lugro-meshhttp://www.lugro.org.ar/mailman/listinfo/nightwing-Bill "another fine mesh" Moffit
Saludos, Julio