Dynamic assignment of primary IP for 802.11s

[Stuart Trusty]

Greetings all,

When we set up a template for something like an 802.11s node, the way I see this is that we must fixate it with a single IP like 192.168.1.1, but this isn't conducive to 802.11s, as each router needs 192.168.1.2, .3, etc. for it to function properly.  

However, the unit can't get on the MAC mesh without knowing its IP in the first place, so using dhcpd in this scenario is a puzzle to me.  

Is anyone using some dynamic or incremental feature in a template to assign IP's on a group of routers, or I am I thinking about going about this in the wrong way?  Clearly Google Wifi has solved this issue, what is the best approach here?

Thank you in advance,

Stuart

[Stuart Trusty]

Hi Federico,

I should mention that I have seen https://github.com/openwisp/django-ipam, https://github.com/openwisp/openwisp-ipam, and this post -> https://groups.google.com/forum/#!topic/openwisp/_TIF0bD8NYA ; you know what I am trying to accomplish-  make it so a 10-year old can set this up- so, this isn't ultimately helpful.  I saw your expo Twitter post, and pulled the PDF, it is very cool, and based on the request therein I am willing to join forces with you; we have a couple of people working on this stuff.  But I need to turn this into something usable for everyone, and if you want to help me coordinate the spec to do this *easily*, I am happy to get it programmed.

Thank you again,

Stuart

[Federico Capoano]

Stuart, an ipam system for what you need to do is overkill.

I also want to make openwisp able to be used by anyone, (not a 10 years old maybe :-P) and I assure you that if I had the resources to do it I would have already done it .. but hey if it was so easy to build a system so complex that is super easy to use and also free of charge someone else would have probably done it earlier than me.

It's cool that you want to join forces but I cannot lie to you and tell you Santa Claus exists, Santa Claus does not exist and this can't be done nor easily nor quickly. If you want to join forces and you have technical people working for you, get them to participate in this community and coordinate with us on specific issues like other contributors have been doing and we'll be able to get some stuff done and improve the system in a way that is useful for you.

Federico

[Stuart Trusty]

Hi Federico!

You know, every time someone says Santa doesn't exist, an elf gets eaten by a reindeer.  Santa does exist!  You just have to tell him what you want.

So, I look at this, from a commercial perspective, and I see this great infrastructure that just needs to sit at a frozen production version and have a different, namely very Simple, UI over just a tiny handful of variables for the user.  

It only will need to manage 802.11s, and we need to standardize on some very cheap radios that anyone can use as far as a supported production version.  I am getting test equipment in from China for both very inexpensive units and some wave-2 based radios.  Then we need to get a nicer looking map to deploy the radios and dump a few stats and be able to change a few things like SSID or captive portal page from a simple control panel.  This is all, really, along with maybe some RADIUS output formatted nicely, to get this mainstream.  What we have here is like Linux itself, and the market wants to do is run an Application.  

If this simple config can become the end-deliverable, it is still OpenWISP, but maybe v3, something simple for everyone to use, and then users can flash whatever radios they want, or they can get the standard issue ones.  I am ready to make our GPS-based mobile-app map work as a control panel for OpenWISP, and I have both an OpenWRT/RADIUS guy, a low-level firmware guy, and a group of core programmers ready to take this in a direction that we are ready to go out and sell.  Maybe we set up a server for a cloud-based application of the simple version and people can pay a nominal fee to help financing some of this development.  I don't think you want to carry this world on your shoulders forever for free, do you?

I do wish you would give some of the technical details that you keep alluding to.  Can we get some real templates for your 802.11s client and gateway, and batman-adv configs appended to this thread, and maybe a short paragraph on the conceptual replacement for ipam for new ip's for newly deployed units?  Now that would Really be helpful for a lot of people, and the whole point here is we don't need everyone re-inventing the wheel.  Right now it is just too complicated, and maybe it is all in your head and that is great, but down here in the trenches, we are not much smarter than the 10-year-olds that this needs to serve in these underdeveloped regions where this technology needs to land.  

Thanks for your consideration, and Ho Ho Ho!

Stuart

[Federico Capoano]

Stuart, I don't know if you read this page yet:

http://openwisp.io/docs/general/values.html

In the goals section it is stated:

Create web interfaces that are easy to use even for people who have limited experience with computer networking concepts (note: we are very far from reaching this goal as of end of 2017)

Bold and red mine.

we are still far from being able to do that. I know it sounds easy to you, but it is not. If you think it's easy, go ahead and build it. I will be happy you have proved me wrong.

Right now I am focused on building the foundation, the core of OpenWISP. I offer my help here on this mailing list for free in my free time, volunteers write documentation for what they are able to do in their free time. I am sorry we don't have configuration templates with 802.11s and batman-adv ready for you. There are many people here using OpenWISP that have probably built something like that and didn't share it nor took the time to document it. I wish they did that but I can't force them to do it.

I cannot stop working on my own priorities to build a batman-adv 802.11s test network and write a tutorial on how to do it because 1. it takes a lot of time to do properly and 2 that is not my priority at the moment, I hope you understand that.

The tones of your mail are inappropriate for an open source community. It seems that you are writing a letter of complaint to the management board of a company in which you are a shareholder. This is not a company. We haven't sold you anything. We are giving away OpenWISP for free of charge under the GPL3 license which states:

The requirement to provide Installation Information does not include a

requirement to continue to provide support service, warranty, or updates

for a work that has been modified or installed by the recipient, or for

the User Product in which it has been modified or installed.

[...]

This program is distributed in the hope that it will be useful,

but WITHOUT ANY WARRANTY; without even the implied warranty of

MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the

GNU General Public License for more details.

If you want to look it for a commercial point of view, I really think you should hire some developers who build what you want for you and interact with us so we can improve the system in a way that allows you to do what you want. There's no point on complaining on this list or asking us to do things for you. It's an open source project and people are supposed to contribute, or pay somebody who does what they want. There's no free lunch here.

Best regards

Federico

--
You received this message because you are subscribed to the Google Groups "OpenWISP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openwisp+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Stuart Trusty]

Hi Federico,  

You might be misunderstanding my tone, as I am certainly not complaining.  I have had my nose in the GPL since 1994, and was one of the earliest supporters of Open-Mesh when it was a free and open project very similar in purpose and values to this one.  I intensely accept the terms of the GPL, and what your are doing here is a welcome relief from the tyranny of Datto Networking that has stolen all of this public code and made it unusable by all of us early developers.  I am ecstatic to have free access to what you have done here, and I feel you are doing the world a great favor by relieving it of the oppression of these wifi portal and network hardware companies that have jacked the consumer cost of Wifi through the roof.  

The point of your project is to serve the needs of the users, and get adoption, right?  I have already reprogrammed an Open-Mesh from scratch once, and if Loren Wolsiffer at WifiGator can find that code, he has offered to jump onboard also.  Even if he doesn't, what you have already done here- the OpenWRT controller- is the heavy lifting that is required; the facelift on top of that was already easy to make once, and the creation of a distribution channel for ready-made hardware, as well as support for that configuration- this is my offer for contribution to your project with a number of programmers already on my staff working towards this very end.  We were just working on this and adding crypto payment gateway to Cloudtrax before Open-Mesh sold us all out.  Now we are here, and are willing to work for you, but we are not doing it as an academic project, we need it to generate revenue from sales to average technical users, we know marketing and sales, and are willing to spend the money to make this happen.  

So, to clarify, no I am not asking you to write any tutorial, I was only asking you to distribute the templates shown in your YouTube video since it would save everyone a lot of time, and if you would kindly consider taking the time to answer the original question, which is what you mean by "an ipam system for what you need to do is overkill" for auto assigning primary router IP's, we can get to work programming all of these things for OpenWISP.  

Thank you for your understanding,
Stuart

You received this message because you are subscribed to a topic in the Google Groups "OpenWISP" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/openwisp/cMMMEQ2d8eo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to openwisp+u...@googlegroups.com.

[Federico Capoano]

Hey Stuart,

On Saturday, November 10, 2018 at 2:00:39 PM UTC+1, Stuart Trusty wrote:

Hi Federico,  

You might be misunderstanding my tone, as

I am certainly not complaining.

Then you need to tone down your emails and focus on technical issues instead of bringing up political / commercial issues which are not the focus of this project nor I believe anyone on this mailing list cares (not everybody who uses openwisp uses it to build a public wifi service and I believe many have never used OpenMesh and some even don't know what we are talking about, many use it as a tool to automate their networks).

 

I have had my nose in the GPL since 1994, and was one of the earliest supporters of Open-Mesh when it was a free and open project very similar in purpose and values to this one.  I intensely accept the terms of the GPL, and what your are doing here is a welcome relief from the tyranny of Datto Networking that has stolen all of this public code and made it unusable by all of us early developers.  I am ecstatic to have free access to what you have done here, and I feel you are doing the world a great favor by relieving it of the oppression of these wifi portal and network hardware companies that have jacked the consumer cost of Wifi through the roof.  

That's great.

 

The point of your project is to serve the needs of the users, and get adoption, right?

The goals of the project are stated here: http://openwisp.io/docs/general/values.html

Adoption will come with time. I'm not interested in rushing because in my experience it is harmful. OpenWISP1 was rushed into existence and once it did what it was designed to do, it couldn't do anything more than that and never became relevant outside of Italy (the country for which it was designed) with the result that many companies and organizations built softwares that were really similar to it but had the same flaws, never achieving any breakthrough.

Once I decided to fix all the issues I have seen and I started to work on NetJSON and OpenWISP 2 I had to do it against the opinion of a lot of people who tried to persuade me it was a bad idea because it would have taken too long and we needed something quickly, "That will take years! We need something now!". Do you think who wanted something now back then has achieved anything like what OpenWISP and NetJSON are achieving today?

The compound effect that this work is going to have in 10 years into the future cannot be estimated. People do not see this, they only focus about building a UI, building a product, but that has been done many times and it has failed many times and I'm not interested in doing it all over again. I'm working on this project for the long term.

You need to understand that if you want to work with OpenWISP.

 

  I have already reprogrammed an Open-Mesh from scratch once, and if Loren Wolsiffer at WifiGator can find that code, he has offered to jump onboard also.  Even if he doesn't, what you have already done here- the OpenWRT controller- is the heavy lifting that is required; the facelift on top of that was already easy to make once, and the creation of a distribution channel for ready-made hardware, as well as support for that configuration- this is my offer for contribution to your project with a number of programmers already on my staff working towards this very end. 

Different people have tried to do the same thing you are doing and have failed. It doesn't mean you will fail as well but I want to set it straight that what you are doing does not represent a contribution to the project.

The project did not improve an inch because of these discussions. The project can only improve with patient work on a feedback loop.

We were just working on this and adding crypto payment gateway to Cloudtrax before Open-Mesh sold us all out.  Now we are here, and are willing to work for you, but we are not doing it as an academic project,

What do you mean to say with "we are not doing it as an academic project"?

This is the kind of off-topic comment that you must avoid because it only fuels confusion. Please stop it.

I understand you are upset because Datto Networks has sent your business plan out of the window but please don't take that frustration here.

I'm sorry OpenWISP doesn't  have the magic wand you need to build your product.

 

we need it to generate revenue from sales to average technical users, we know marketing and sales, and are willing to spend the money to make this happen.  

 

Throwing money at the problem is not the solution.

If you want to work with OpenWISP, you and your developers simply need to work patiently with us, follow the conventions the project is already following, avoid bringing up issues which are not technical or off-topic on this list (if you really want to do that you should take the discussion off-list).

If you don't want to do that I highly advice you to get a commercial agreement with some other company building a similar product.

So, to clarify, no I am not asking you to write any tutorial, I was only asking you to distribute the templates shown in your YouTube video since it would save everyone a lot of time, and if you would kindly consider taking the time to answer the original question, which is what you mean by "an ipam system for what you need to do is overkill" for auto assigning primary router IP's, we can get to work programming all of these things for OpenWISP.  

Great! So let's stop bringing up all the other issues about datto networks, open mesh, crypto, how the current OpenWISP UI is not good enough for 10 years olds, that you're not building an academic project (totally inappropriate and out of the blue comment that you must learn to save for yourself) and focus on the technical questions and I'm sure next communications will go smoother.

An IPAM is a system used by organizations which manage big networks to assign IP, subnets and vlans to different departments, buildings, segments of their data center and so on and for what you described it is overkill. Moreover the IPAM module of OpenWISP is not ready for production usage nor released yet, so it's targeted only at developers who have the skills and patience to help us improve it gradually as we understand more and more how to make it better and integrate it properly with the rest of OpenWISP.

If you want to autoconfigure IP addresses in a mesh network there are other solutions for that which you should better ask in the forums / mailing list of the tools you're using to build the mesh network, so if you are using batman-advanced, I suggest you to ask this question in that community instead of here because there are more experts about batman-adv there than here and you have a higher chance of finding what you want.

There's also another popular mesh solution which is meant to be  easy to use for everybody: https://libremesh.org/ - but if you write to their list, I highly suggest you to focus on the technical aspects and avoid any other not-relevant comment.

The templates I have shown in my youtube videos and elsewhere in my presentations are already shared in the documentation of the OpenWRT configuration backend (http://netjsonconfig.openwisp.org/en/latest/backends/openwrt.html), I was using only layer2 802.11s with no batman-adv.

Whenever I needed to learn how to implement a use case in OpenWISP, whether it was a public wifi bridged to a layer2 VPN with OpenVPN, a mesh network with 802.11s, WPA2 Enterprise for eduroam, I dug up, looked on google, asked advice to other people in the OpenWRT community, did a lot of test, find out the configuration that worked and I've documented all I could, infact you can find all these things I'm citing in the netjsonconfig documentation.

I suggest you to do the same: do your own research to implement what you need and publish your findings. If you don't like the tooling with which the OpenWISP documentaton is built, publish your findings anywhere and in any way you prefer. Somebody else will help us to include your findings in the official documentation.

Federico