Hello all together,
Here in Austria we use our 44.143/16 net purely for Hamnet/Intranet purpose. Most users are not on this mailing list and simple use the network.
Most of the network is radio based with a few interlinks tunneled trough the internet
https://hamnetdb.net/map.cgi#zoom=8&lat=47.811&lon=13.37&layer=M...
The nodes are using BGP routing as most of Hamnet in Europe, most users are only connecting to the Network and don't have to care about BGP.
The question I'm getting over and over again is, which Subnets should be routed to Hamnet to reach most of it and at the same time not blocking the Internet routed subnets. So this clear separation would clearly help us as well.
Finally, those OMs who have Internet routed Subnets (outside 44.143) would have to move, but its clearly worth the effort. Some of us are in training as we had to move allot of infrastructure after the subnet was sold.
73 Lucas OE2LSP/KC1GDN
Some remarks following
Am 28.07.21 um 21:27 schrieb Lucas Speckbacher OE2LSP via 44Net:
... Here in Austria we use our 44.143/16 net purely for Hamnet/Intranet purpose. Most users are not on this mailing list and simple use the network.
I think the word "Intranet" somehow was misunderstood in this long discussion. To point it out a little bit clearer:
Hamnet/intranet never was intended to be a fully closed network. Obviously all hamnet-users want to talk with every other 44net-user around the world and vice versa without any form of nat because this could block some wanted use-cases. NAT is evil!
Therefor it is indispensable to keep those networks within the 44net-IP-range! Just to keep things simple and stupid. 1:1 mapping is no option. It wouldn't save a single IP on the 44net-side.
... The question I'm getting over and over again is, which Subnets should be routed to Hamnet to reach most of it and at the same time not blocking the Internet routed subnets. So this clear separation would clearly help us as well.
This is one of the things to get solved and TAC-proposal is aiming at. One of possible solutions could be: renumbering for a minority of networks giving them a lot of time and support to do that.
In this long lasting thread I never ever have read that this would be technically impossible. Instead I read a lot of something like: "Baaahh, that is a lot of work" etc
Antonious pointed it out again and again why it could be better to get more structure in the net as a first step. I dispence in repeating all that here again. Logically it is the second step to roll out a technical solution to keep everyone satisfied.
That is when Rob and others could enter the game. A PoP-Structure around the world looks very interesting for me. Antonious has made it clear that TAC is working on a solution like that. Hopefully this doesn't last another 8 months because there has a lot of work been done and published on this point ;-)
Finally, those OMs who have Internet routed Subnets (outside 44.143) would have to move, but its clearly worth the effort.
ACK
Some of us are in training as we had to move a lot of infrastructure > after the subnet was sold.> 73 Lucas OE2LSP/KC1GDN
73s de DD9QP
Finally, those OMs who have Internet routed Subnets (outside 44.143) would have to move, but its clearly worth the effort.
ACK
I disagree, I'm after reading the proposal carefully front to back for the third time today. The more I read it the more I'm coming to the conclusion is that it seems to be un-necessary work being thrust upon everyone in the 44.128/10 space to renumber their networks just to have the exact same feature set and (now thoroughly annoyed) users after all the effort.
This doesn't actually make any sense to me now.
Surely we can be more ambitious and do better than that?
John
EI7IG
I would agree, I'm still trying to wrap my head around what real technical problem is renumber trying to resolve, and I can't help but keeping coming back to the conclusion, the problem trying to be solved is not that of an immediate technical nature, but that of a matter of convenience. If all of the publicly routable IP's are moved into 1 block, the other block becomes fair game for reallocation in the future. This reallocation could be for either, network expansion of the routable space, or offered for sale. I understand that the statements are put in place saying we are not selling, But to be fair that was what was said a couple of years back as well, and now we have a /10 less of space.
On Wed, Jul 28, 2021 at 1:41 PM John Ronan via 44Net 44net@mailman.ampr.org wrote:
Finally, those OMs who have Internet routed Subnets (outside 44.143) would have to move, but its clearly worth the effort.
ACK
I disagree, I'm after reading the proposal carefully front to back for the third time today. The more I read it the more I'm coming to the conclusion is that it seems to be un-necessary work being thrust upon everyone in the 44.128/10 space to renumber their networks just to have the exact same feature set and (now thoroughly annoyed) users after all the effort.
This doesn't actually make any sense to me now.
Surely we can be more ambitious and do better than that?
John
EI7IG
44Net mailing list 44Net@mailman.ampr.org https://mailman.ampr.org/mailman/listinfo/44net
Not only that, fully half of the networks in .US would have to move. Those states that were unlucky enough to have been numbered above 44.63, in order to clear 44.64/10 from active assignment. This includes 44.127 “multinational” block, which includes the /20 IRLP uses, assigned just a few months ago. It would be fairly easy for me to move to a new block below 44.63, but I have to ask why? I am guessing other folks would have a much more difficult time.
Dave K9DC
On Jul 28, 2021, at 16:41, John Ronan via 44Net 44net@mailman.ampr.org wrote:
Finally, those OMs who have Internet routed Subnets (outside 44.143) would have to move, but its clearly worth the effort.
ACK
I disagree, I'm after reading the proposal carefully front to back for the third time today. The more I read it the more I'm coming to the conclusion is that it seems to be un-necessary work being thrust upon everyone in the 44.128/10 space to renumber their networks just to have the exact same feature set and (now thoroughly annoyed) users after all the effort. This doesn't actually make any sense to me now. Surely we can be more ambitious and do better than that?
John EI7IG
Hello Dave! This current proposal does not renumber anyone between 44.64 and 44.127 (inclusive) unless they want to move to 44.128/10. The exact wording is that we leave all existing allocations there, and we simply don’t make new ones, in the future.
I hope this clarifies the proposal a little bit.
Antonis
On 28 Jul 2021, at 23:14, Dave Gingrich via 44Net 44net@mailman.ampr.org wrote:
Not only that, fully half of the networks in .US would have to move. Those states that were unlucky enough to have been numbered above 44.63, in order to clear 44.64/10 from active assignment. This includes 44.127 “multinational” block, which includes the /20 IRLP uses, assigned just a few months ago. It would be fairly easy for me to move to a new block below 44.63, but I have to ask why? I am guessing other folks would have a much more difficult time.
Dave K9DC
Hello all,
I administer a very small area of the "HamNet" network in Northern Germany. I have 11 HamNet nodes here, which are connected via WiFi directional radio links on amateur radio frequencies. The routing is done via BGP and the users access this network also via directional antennas on the roof of their houses.
This only works when each user sets a a static route to the antenna on their router. This used to be 44.0.0.0/8, but due to the constant fragmentation of the network, there are now more than 4 routes, which does not make it easy for the end users and the administrators. Incidentally, this is also the reason why I only registered on this mailing list today: so far, 44.0.0.0/8 is still configured in my router, i.e. all 44-services on the internet are not accessible for me.
Could someone briefly tell me which subnets I can only reach via the internet and which only via amateur radio? I would like to update my route settings.
I keep reading the question here why we are setting up "our own separate intranet" here in europe. The answer is: Because we can.
Because our amateur radio licence allows us to experiment with radio technology on our amateur radio frequencies. Because we are not only users in this network, but can also develop and offer services ourselves. For example, we connecct digital and analogue FM repeaters via our own network. We don't depend on an internet connection or mobile phone providers. We do it ourselves. In the same way, we offer remote stations. Here, I don't have to secure access against criminal activities, because only radio amateurs have access to that network. Messages and telephone calls? They also run over our self-made network on amateur radio frequencies.
We no longer build the radios ourselves, but we do build the radio links and applications ontop of it. I can only do this as a radio amateur and of course I use the 44 network assigned to me.
But this only works as long as I can set a defined route for my antenna.
73 de Gerrit, DH8GHH
Thank you for that clarification. I missed that.
I assume there are fewer users in 44.128/10 then on hamnet and i assume that's the reason why the 44.128/10 users are being asked to renumber. Where i think the elephant in the room is that the hamnet users should not be using public routed capable address space in the first place. There is no technical reason why they should use public addresses and not RFC 6598 for example or RFC 1918. In this case RFC 6598 is more appropriate i think as it is being natted in places onto the internet and provides a type of utility.
My day job is in carrier internet which i have been involved with since the early 90's. I have been evolved in everything from policy making to nuts and bolts core network design I even have my name in some of the code that you likely use every day as part of the internet backbone and if I told any of my peers a bunch of hobbits were using millions of dollars of publicly addressable internet ranges for an intranet I would struggle to convince them i was not jokeing.
The excuse "Because we can" to squander millions of dollars worth of resource is very hard to justify.
Now dont get me wrong im very grateful to have the resource and I will renumber my BGP advertised to the public internet direct 44.128/10 range if that's the consensus. But I think its very hard to justify using any space at all for an intranet in 2021. Carriers and ISP's and enterprises have spent that last 10 years and millions of dollars solving that exact problem and the answer is IPv6 and RFC6598
73
Hello Matt, thanks a lot for the message!
On 30 Jul 2021, at 13:11, Matt Perkins via 44Net 44net@mailman.ampr.org wrote:
I assume there are fewer users in 44.128/10 then on hamnet and i assume that's the reason why the 44.128/10 users are being asked to renumber. Where i think the elephant in the room is that the hamnet users should not be using public routed capable address space in the first place. There is no technical reason why they should use public addresses and not RFC 6598 for example or RFC 1918. In this case RFC 6598 is more appropriate i think as it is being natted in places onto the internet and provides a type of utility.
My day job is in carrier internet which i have been involved with since the early 90's. I have been evolved in everything from policy making to nuts and bolts core network design I even have my name in some of the code that you likely use every day as part of the internet backbone and if I told any of my peers a bunch of hobbits were using millions of dollars of publicly addressable internet ranges for an intranet I would struggle to convince them i was not jokeing.
The excuse "Because we can" to squander millions of dollars worth of resource is very hard to justify.
Now dont get me wrong im very grateful to have the resource and I will renumber my BGP advertised to the public internet direct 44.128/10 range if that's the consensus. But I think its very hard to justify using any space at all for an intranet in 2021. Carriers and ISP's and enterprises have spent that last 10 years and millions of dollars solving that exact problem and the answer is IPv6 and RFC6598
Unfortunately, as I said in previous e-mails, we cannot use a space such as RFC1918 or Carrier Grade NAT because it is not globally unique. There are ISPs that use 10/8 addresses that are assigned to end users in their WAN, and there are ISPs that use RFC6598 for their customers’ WAN connections too. So there will be overlaps, conflicts, and problems. We absolutely have to ensure global uniqueness and guarantee non-overlap for this to work at a large scale. And the only way we know how to do this is with space from 44/8.
Again, going back to the similar analogy of RF spectrum: Why should we give these “radio amateurs” so much and so valuable spectrum? Why waste literal billions worth of it globally? If they want to communicate, they should use the phone system or the Internet like everything else. I don’t see a reason for them to get this tremendously expensive resource “because they can” or “because they want to”.
Last December, Greece, a tiny country with only 10M people proceeded to rent spectrum for 5G. They rented this spectrum for 15 years (not forever) and only in a few bands (not across the entire thing). Just for one tiny country, just for 15 years, just for one technology (5G) the government made 372M EUR or close to $450M. This is more money than the entire 44/8. If we add up the cost of the entire 5G spectrum around the world, we are clearly in the billions of dollars.
They gave a 15-year license for parts of: 3400-3800 MHz, 700 MHz, 2 GHz, and 26 GHz.
Today radio amateurs have “for free” so many more bands, so many more allocations, so much more spectrum and bandwidth. And why? One could argue that 5G will serve much more people and much more important use cases than ham radio spectrum. One could also argue that a /24 on 44 net provides less to the society as a whole than a /24 in AWS or anyone else.
Should we return all our spectrum back because we “waste millions of dollars worth of resource”? I think we’re “wasting” not just millions with spectrum but probably tens of billions.
Would we be happy if the FCC told our US friends that they’re taking back all spectrum allocated to amateur radio and they can now use the frequencies of WiFi or FM (Music) Radio if they want, as secondary users, and never cause any interference or problems to primary users (stations or WiFi or BT devices or microwave ovens)? Technically if they do this, they don’t “waste” billions of dollars of space, and they can push these “intranet” people to a shared medium. Let them then figure out an ISM band where they don’t overlap with any other use case in this local area. Take back the dedicated resources.
—
As you can tell, there is a large argument to be made for using 44.128/10 for the ham-to-ham communication. The reason we want radio amateurs to use this space is because it’s theirs, and they can do whatever they want with it. They got it allocated, so they can use it. Hopefully, unlike the real RF spectrum, it won’t be taken back from them if they don’t…
Antonis
Hello fellows ampr users,
Just sharing analysis on ampr actual portal registrered figures:
U.S. address space 44.0/9 = 8M possible hosts = 66% Rest of the world 44.128/10 = 4M possible hosts = 33%
Actual allocations on http://amsat.org.ar/amprhosts.txt 41,150 total allocations 100% Total, on which 7,480 U.S. allocations 18% 33,670 Rest of the world 82%
Conclusion: US has the larger space available and fewer users Rest of World, has a smaller space and most users
Thus rest of the world has activity and thus experience on amprnet.
Furthermore the proposed 'PoP' (Point of Presence if that is what was implied) will be hard and costly to implement for ISPs specially on smaller/remote countries, wonder if local ISPs will do and if ARDC will fund PoPs on the many ISPs providing conectivity.
Should had been wiser if amprnet was heard/polled before proposing to ARDC.
73, lu7abf, Pedro Converso 44.153 Argentina Coordination
On 7/30/21, Antonios Chariton (daknob) via 44Net 44net@mailman.ampr.org wrote:
Hello Matt, thanks a lot for the message!
On 30 Jul 2021, at 13:11, Matt Perkins via 44Net 44net@mailman.ampr.org wrote:
I assume there are fewer users in 44.128/10 then on hamnet and i assume that's the reason why the 44.128/10 users are being asked to renumber. Where i think the elephant in the room is that the hamnet users should not be using public routed capable address space in the first place. There is no technical reason why they should use public addresses and not RFC 6598 for example or RFC 1918. In this case RFC 6598 is more appropriate i think as it is being natted in places onto the internet and provides a type of utility.
My day job is in carrier internet which i have been involved with since the early 90's. I have been evolved in everything from policy making to nuts and bolts core network design I even have my name in some of the code that you likely use every day as part of the internet backbone and if I told any of my peers a bunch of hobbits were using millions of dollars of publicly addressable internet ranges for an intranet I would struggle to convince them i was not jokeing.
The excuse "Because we can" to squander millions of dollars worth of resource is very hard to justify.
Now dont get me wrong im very grateful to have the resource and I will renumber my BGP advertised to the public internet direct 44.128/10 range if that's the consensus. But I think its very hard to justify using any space at all for an intranet in 2021. Carriers and ISP's and enterprises have spent that last 10 years and millions of dollars solving that exact problem and the answer is IPv6 and RFC6598
Unfortunately, as I said in previous e-mails, we cannot use a space such as RFC1918 or Carrier Grade NAT because it is not globally unique. There are ISPs that use 10/8 addresses that are assigned to end users in their WAN, and there are ISPs that use RFC6598 for their customers’ WAN connections too. So there will be overlaps, conflicts, and problems. We absolutely have to ensure global uniqueness and guarantee non-overlap for this to work at a large scale. And the only way we know how to do this is with space from 44/8.
Again, going back to the similar analogy of RF spectrum: Why should we give these “radio amateurs” so much and so valuable spectrum? Why waste literal billions worth of it globally? If they want to communicate, they should use the phone system or the Internet like everything else. I don’t see a reason for them to get this tremendously expensive resource “because they can” or “because they want to”.
Last December, Greece, a tiny country with only 10M people proceeded to rent spectrum for 5G. They rented this spectrum for 15 years (not forever) and only in a few bands (not across the entire thing). Just for one tiny country, just for 15 years, just for one technology (5G) the government made 372M EUR or close to $450M. This is more money than the entire 44/8. If we add up the cost of the entire 5G spectrum around the world, we are clearly in the billions of dollars.
They gave a 15-year license for parts of: 3400-3800 MHz, 700 MHz, 2 GHz, and 26 GHz.
Today radio amateurs have “for free” so many more bands, so many more allocations, so much more spectrum and bandwidth. And why? One could argue that 5G will serve much more people and much more important use cases than ham radio spectrum. One could also argue that a /24 on 44 net provides less to the society as a whole than a /24 in AWS or anyone else.
Should we return all our spectrum back because we “waste millions of dollars worth of resource”? I think we’re “wasting” not just millions with spectrum but probably tens of billions.
Would we be happy if the FCC told our US friends that they’re taking back all spectrum allocated to amateur radio and they can now use the frequencies of WiFi or FM (Music) Radio if they want, as secondary users, and never cause any interference or problems to primary users (stations or WiFi or BT devices or microwave ovens)? Technically if they do this, they don’t “waste” billions of dollars of space, and they can push these “intranet” people to a shared medium. Let them then figure out an ISM band where they don’t overlap with any other use case in this local area. Take back the dedicated resources.
—
As you can tell, there is a large argument to be made for using 44.128/10 for the ham-to-ham communication. The reason we want radio amateurs to use this space is because it’s theirs, and they can do whatever they want with it. They got it allocated, so they can use it. Hopefully, unlike the real RF spectrum, it won’t be taken back from them if they don’t…
Antonis _________________________________________ 44Net mailing list 44Net@mailman.ampr.org https://mailman.ampr.org/mailman/listinfo/44net
-
Antonios Chariton (daknob) -
Darcy Buskermolen -
Dave Gingrich -
DD9QP Egbert -
Gerrit Herzig DH8GHH -
John Ronan -
Lucas Speckbacher OE2LSP -
Matt Perkins -
Pedro Converso