One idea for a framework however would be to use something like Consul to do service discovery and post it on said website/directory.

Consul look very interesting.

https://www.consul.io/

Thanks for the post!

Bill

On 3/8/2016 7:01 PM, Leon D. Zetekoff wrote:

(Please trim inclusions from previous messages) _______________________________________________

On 3/8/2016 11:09 AM, Don Fanning wrote:

...

(Please trim inclusions from previous messages) _______________________________________________ On Tue, Mar 8, 2016 at 3:50 AM, marius@yo2loj.ro wrote:

...

(Please trim inclusions from previous messages) _______________________________________________ Just that there will be alot of opposition from people worrying about 1 ping an hour should some multicast group announcement and subscription messages appear on the network...

Ultimately that is the biggest problem. We can't even get network operators to allow ICMP packets without them scrutinizing every packet that comes across the wire as a threat. Among other topics...

+1 on that. I am a network engineer and worked for carriers and it really ticks me off when pings are turned off. Pings are the lowest priority traffic on any network. not responding to pings makes it that much more difficult to debug when things go wrong.

ALso, if folks are so worried where pings are coming from ACLs can be crafted IF necessary to keep unwanted traffic out.

Also, why are we reinventing the wheel? There are already protocols and tools to do all this. It should be automatic like BGP and OSPF are. ALso, Marius' clever RIP scripts for Mikrotik work really well (after we debugged and finetuned it)

There's no real reason to reinvent anything. We are still in the 1990s network wise and we need to jump the shark and get into this century. While I am not (and never was) in favor of netrom and all that stuff back in the day I was one of the rebels in South FLorida using Phils KA9Q NOS.

I know there are some good implementations of integrating all the legacy stuff and UROnode comes to mind. But do we have to still use this in 2016? can't we migrate everything to tcpip?

73 leon

Leon D. Zetekoff, NCE, MTCE

WA4ZLW

BackWoods Wireless

Blandon, PA 19510

Facebook http://s.wisestamp.com/links?url=http%3A%2F%2Fwww.facebook.com%2Fprofile.php%3Fid%3Dleonzetekoff LinkedIn http://s.wisestamp.com/links?url=http%3A%2F%2Fwww.linkedin.com%2Fin%2Fleonzetekoff http://s.wisestamp.com/links?url=http%3A%2F%2Fwww.linkedin.com%2Fin%2Fleonzetekoff Get a signature like this. http://s.wisestamp.com/links?url=http%3A%2F%2Fr1.wisestamp.com%2Fr%2Flanding%3Fpromo%3D17%26dest%3Dhttp%253A%252F%252Fwww.wisestamp.com%252Femail-install%253Futm_source%253Dextension%2526utm_medium%253Demail%2526utm_campaign%253Dpromo_17 CLICK HERE. http://s.wisestamp.com/links?url=http%3A%2F%2Fr1.wisestamp.com%2Fr%2Flanding%3Fpromo%3D17%26dest%3Dhttp%253A%252F%252Fwww.wisestamp.com%252Femail-install%253Futm_source%253Dextension%2526utm_medium%253Demail%2526utm_campaign%253Dpromo_17

When I now start to re alive our amprnet the main goal was to treat the network as a Backup for "other Networks" for emergencies as example (and because of it i plan to make a backbone high speed country wide Wifi Based in order not to be depended on any wire connection) I looked on the AMPRNET network as a "HF network" and then i discovered that the network not act as a real mesh ... I can not connect from my ampr net to certain ampr nets Now lets take a case that the hams community in Thailand have a disaster and we (the rest of the ham community ) want to help them ... but unfortunately we have no contact to their network ... and why ? because we use "old technologies" like ENCAP TABLE or even worse because the Gateway operator didnt put the whole encap table in his gateway (from his own reasons) All over the world the Governments (but not only them) looked on the ham radios as peoples that in communication disaster can make contact with their radios , unfortunately i can not say it on the amprnet even in non disaster situations Currently the only advantage of the AMPRNET as i see it is the big address space we have that we can transfer on a single IP that are transparent and can be accessed from the outside world but thats not what i would like to see the AMPRNET I would like to see it be able to be connected like Net/Rom can do or Like OSPF or any other "smart routing protocols" then I know that in a "real situation" i will be able to gain access any part of the AMPR network because the Network itself take care of its connectivity and as long as the specific network have any kind of connection to the outside world (RF or Wire) the network will be able to reach it I know we can do it we have in our community the world best experts in their field and a huge knowledge and good willing Im willing to help as needed in any subject that i can .... Best Regards Ronen - 4Z4ZQ http://www.ronen.org

________________________________________ From: 44Net 44net-bounces+ronenp=hotmail.com@hamradio.ucsd.edu on behalf of Leon D. Zetekoff wa4zlw@backwoodswireless.net Sent: Tuesday, March 8, 2016 4:08 PM To: 44net@hamradio.ucsd.edu Subject: Re: [44net] Is there raceroutre machine on 44 net available for public ?

(Please trim inclusions from previous messages) _______________________________________________

. We are still in the 1990s network wise and we need to jump the shark and get into this century.

Why we use IPIP and a full mesh?

Because IPIP is the simplest tunneling protocol, and the only IPv4 tunnel with available point to multipoint support on some systems (which simplifies the setup a lot, unfortunately routers do not support it). And because a full mesh is the only architecture tolerating arbitrary gateway breakdowns without affecting other gateways/subnets, and it does not have a single point of failure.

All of you want BGP and OSPF. Nice. And how do you implement that in a full mesh fashion? Where do you get the mesh peer information from? The answer is simple, either from a central server like the one we have now, offering something like the ipencap file for static configs, the RIP broadcasts for dynamic configuration and the portal API for scripting purposes, or you can drop the mesh concept, falling back to standard gateway architectures like the classic client/ISP level internet setup.

For a mesh this information gathering step can not be circumvented. Even if we change routing protocols, you still need to know your peers BEFORE being able to set up BGP or OSPF. And if you know your peers, you can know their subnets, too. So there is actually no need for a routing protocol running on a limited bandwidth network, because the information is already there, in your peering data.

Of course, no one stops you from privately peering with other systems and using these protocols (I do...). But there is no real need to do it.

Marius, YO2LOJ

-----Original Message----- From: f1sca Sent: Wednesday, March 09, 2016 10:03 To: 'AMPRNet working group' Subject: Re: [44net] Is there raceroutre machine on 44 net available forpublic ?

(Please trim inclusions from previous messages) _______________________________________________ Totally agree with you Ronen. Our AMPRNET should be designed and work as Internet, on waves. Tunneling should be considered As a work around of lack of radio link. Why do we need to steel use IPIP ? Can't we try to test other routing schemes ?

Regards

F1SCA

-----Message d'origine----- De : 44Net [mailto:44net-bounces+f1sca=numericable.fr@hamradio.ucsd.edu] De la part de R P Envoyé : mercredi 9 mars 2016 08:28 À : AMPRNet working group 44net@hamradio.ucsd.edu Objet : Re: [44net] Is there raceroutre machine on 44 net available for public ?

(Please trim inclusions from previous messages) _______________________________________________ When I now start to re alive our amprnet the main goal was to treat the network as a Backup for "other Networks" for emergencies as example (and because of it i plan to make a backbone high speed country wide Wifi Based in order not to be depended on any wire connection) I looked on the AMPRNET network as a "HF network" and then i discovered that the network not act as a real mesh ... I can not connect from my ampr net to certain ampr nets Now lets take a case that the hams community in Thailand have a disaster and we (the rest of the ham community ) want to help them ... but unfortunately we have no contact to their network ... and why ? because we use "old technologies" like ENCAP TABLE or even worse because the Gateway operator didnt put the whole encap table in his gateway (from his own reasons) All over the world the Governments (but not only them) looked on the ham radios as peoples that in communication disaster can make contact with their radios , unfortunately i can not say it on the amprnet even in non disaster situations Currently the only advantage of the AMPRNET as i see it is the big address space we have that we can transfer on a single IP that are transparent and can be accessed from the outside world but thats not what i would like to see the AMPRNET I would like to see it be able to be connected like Net/Rom can do or Like OSPF or any other "smart routing protocols" then I know that in a "real situation" i will be able to gain access any part of the AMPR network because the Network itself take care of its connectivity and as long as the specific network have any kind of connection to the outside world (RF or Wire) the network will be able to reach it I know we can do it we have in our community the world best experts in their field and a huge knowledge and good willing Im willing to help as needed in any subject that i can .... Best Regards Ronen - 4Z4ZQ http://www.ronen.org

________________________________________ From: 44Net 44net-bounces+ronenp=hotmail.com@hamradio.ucsd.edu on behalf of Leon D. Zetekoff wa4zlw@backwoodswireless.net Sent: Tuesday, March 8, 2016 4:08 PM To: 44net@hamradio.ucsd.edu Subject: Re: [44net] Is there raceroutre machine on 44 net available for public ?

(Please trim inclusions from previous messages) _______________________________________________

. We are still in the 1990s network wise and we need to jump the shark and get into this century. _________________________________________ 44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net

_________________________________________ 44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net

MARius makes a good point. A few months ago I didn't like the use of RIP but the ip encapsulation also is a low overhead tunnel. I've got OSPF working here and I can inject the AMPRNet routes in there IF I want onto my private non-AMPR lan. RIght now I don't and I gateway through my router. My watchguard firewall handles all internet traffic EXCEPTING my own AMPRnet --> Internet I gateway it out through my watchguard doing a NAT I believe. It works

Marius - would you like to do some OSPF?

Leon

On 3/9/2016 3:33 AM, Marius Petrescu wrote:

(Please trim inclusions from previous messages) _______________________________________________ Why we use IPIP and a full mesh?

Because IPIP is the simplest tunneling protocol, and the only IPv4 tunnel with available point to multipoint support on some systems (which simplifies the setup a lot, unfortunately routers do not support it). And because a full mesh is the only architecture tolerating arbitrary gateway breakdowns without affecting other gateways/subnets, and it does not have a single point of failure.

All of you want BGP and OSPF. Nice. And how do you implement that in a full mesh fashion? Where do you get the mesh peer information from? The answer is simple, either from a central server like the one we have now, offering something like the ipencap file for static configs, the RIP broadcasts for dynamic configuration and the portal API for scripting purposes, or you can drop the mesh concept, falling back to standard gateway architectures like the classic client/ISP level internet setup.

For a mesh this information gathering step can not be circumvented. Even if we change routing protocols, you still need to know your peers BEFORE being able to set up BGP or OSPF. And if you know your peers, you can know their subnets, too. So there is actually no need for a routing protocol running on a limited bandwidth network, because the information is already there, in your peering data.

Of course, no one stops you from privately peering with other systems and using these protocols (I do...). But there is no real need to do it.

Marius, YO2LOJ

-----Original Message----- From: f1sca Sent: Wednesday, March 09, 2016 10:03 To: 'AMPRNet working group' Subject: Re: [44net] Is there raceroutre machine on 44 net available forpublic ?

(Please trim inclusions from previous messages) _______________________________________________ Totally agree with you Ronen. Our AMPRNET should be designed and work as Internet, on waves. Tunneling should be considered As a work around of lack of radio link. Why do we need to steel use IPIP ? Can't we try to test other routing schemes ?

Regards

F1SCA

-----Message d'origine----- De : 44Net [mailto:44net-bounces+f1sca=numericable.fr@hamradio.ucsd.edu] De la part de R P Envoyé : mercredi 9 mars 2016 08:28 À : AMPRNet working group 44net@hamradio.ucsd.edu Objet : Re: [44net] Is there raceroutre machine on 44 net available for public ?

(Please trim inclusions from previous messages) _______________________________________________ When I now start to re alive our amprnet the main goal was to treat the network as a Backup for "other Networks" for emergencies as example (and because of it i plan to make a backbone high speed country wide Wifi Based in order not to be depended on any wire connection) I looked on the AMPRNET network as a "HF network" and then i discovered that the network not act as a real mesh ... I can not connect from my ampr net to certain ampr nets Now lets take a case that the hams community in Thailand have a disaster and we (the rest of the ham community ) want to help them ... but unfortunately we have no contact to their network ... and why ? because we use "old technologies" like ENCAP TABLE or even worse because the Gateway operator didnt put the whole encap table in his gateway (from his own reasons) All over the world the Governments (but not only them) looked on the ham radios as peoples that in communication disaster can make contact with their radios , unfortunately i can not say it on the amprnet even in non disaster situations Currently the only advantage of the AMPRNET as i see it is the big address space we have that we can transfer on a single IP that are transparent and can be accessed from the outside world but thats not what i would like to see the AMPRNET I would like to see it be able to be connected like Net/Rom can do or Like OSPF or any other "smart routing protocols" then I know that in a "real situation" i will be able to gain access any part of the AMPR network because the Network itself take care of its connectivity and as long as the specific network have any kind of connection to the outside world (RF or Wire) the network will be able to reach it I know we can do it we have in our community the world best experts in their field and a huge knowledge and good willing Im willing to help as needed in any subject that i can .... Best Regards Ronen - 4Z4ZQ http://www.ronen.org

---

From: 44Net 44net-bounces+ronenp=hotmail.com@hamradio.ucsd.edu on behalf of Leon D. Zetekoff wa4zlw@backwoodswireless.net Sent: Tuesday, March 8, 2016 4:08 PM To: 44net@hamradio.ucsd.edu Subject: Re: [44net] Is there raceroutre machine on 44 net available for public ?

(Please trim inclusions from previous messages) _______________________________________________

. We are still in the 1990s network wise and we need to jump the shark and get into this century. _________________________________________ 44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net

---

44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net _________________________________________ 44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net

Brian...yeah wrong quote - senior moment. But we do need to move more into this century. You guys have done a good job with what we have we just need to get more of it out there.

leon

On 3/9/2016 5:40 AM, Brian Kantor wrote:

(Please trim inclusions from previous messages) _______________________________________________

...

. We are still in the 1990s network wise and we need to jump the shark and get into this century.

Excuse me, but "jump the shark" means to make a colossal mistake that signals the eventual decline of something. Certainly that's not what you consciously meant.

• Brian

---

44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net

Totally agree !!!

________________________________________ From: 44Net 44net-bounces+ronenp=hotmail.com@hamradio.ucsd.edu on behalf of Leon D. Zetekoff wa4zlw@backwoodswireless.net Sent: Tuesday, March 8, 2016 4:01 PM To: 44net@hamradio.ucsd.edu Subject: Re: [44net] Is there raceroutre machine on 44 net available for public ?

(Please trim inclusions from previous messages) _______________________________________________

not responding to pings makes it that much more difficult to debug when things go wrong.

Leon