Is it just me or we have a error in the gateway routing table?
entry: 44.9.0.1/255.255.255.252 via 207.111.203.194
This gives a invalid subnet/netmak pair and gets ignored by rip44d.
Should be either 44.9.0.1/255.255.255.255 or 44.9.0.0/255.255.255.252
73 de Marius, YO2LOJ
On 2/22/2012 2:33 PM, Marius Petrescu wrote:
Is it just me or we have a error in the gateway routing table?
entry: 44.9.0.1/255.255.255.252 via 207.111.203.194
This gives a invalid subnet/netmak pair and gets ignored by rip44d.
Should be either 44.9.0.1/255.255.255.255 or 44.9.0.0/255.255.255.252
It's my error, soon to be fixed.
Also, just got an email from the coordinator saying the 44.9.0 addresses were in error, so the whole entry will go away and come back different.
73 de David WA6NMF
73 de Marius, YO2LOJ
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net
On 2/22/2012 3:08 PM, Brian Kantor wrote:
On Thu, Feb 23, 2012 at 12:33:32AM +0200, Marius Petrescu wrote:
Should be either 44.9.0.1/255.255.255.255 or 44.9.0.0/255.255.255.252
44.9 is an invalid subnet.
Dan K6DLC says his assignment of these addresses to me was in error, new addresses are in 44.4.9, new gateway entry in the works.
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can hope!)
73 de David WA6NMF
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF wa6nmf@josephson.com wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can hope!)
I think the reason we started it that way was the sparse population of addresses and the fact we coordinators didn't initially know any better coupled with the geographical/topological distribution of IP nodes where we couldn't really count on a node being in any specific location within the net. Nodes had to determine their neighbors by discovery and they were routed manually.
I didn't start subnetting until users wanted blocks of IP addresses for specific purposes, like UHF vs VHF gateways, digipeaters, or ARES/RACES. I wrote a paper on it but I don't know how widely it was distributed or how well it was received.
Geoff, Would you mind sending a copy of your paper to the group?
K6DLC
On 2/22/2012 6:47 PM, Geoff Joy wrote:
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF wa6nmf@josephson.com wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can hope!)
I think the reason we started it that way was the sparse population of addresses and the fact we coordinators didn't initially know any better coupled with the geographical/topological distribution of IP nodes where we couldn't really count on a node being in any specific location within the net. Nodes had to determine their neighbors by discovery and they were routed manually.
I didn't start subnetting until users wanted blocks of IP addresses for specific purposes, like UHF vs VHF gateways, digipeaters, or ARES/RACES. I wrote a paper on it but I don't know how widely it was distributed or how well it was received.
Hi Geoff, I am in the process of revamping and reassigning ampr addresses in the Greater Los Angeles area (Orange, Los Angeles, Ventura and Santa Barbara counties) and would love to look over your paper. Thanks Don WB5EKU
On Wed, Feb 22, 2012 at 6:54 PM, Daniel Curry daniel@danielcurry.net wrote:
Geoff, Would you mind sending a copy of your paper to the group?
K6DLC
On 2/22/2012 6:47 PM, Geoff Joy wrote:
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF wa6nmf@josephson.com wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can hope!)
I think the reason we started it that way was the sparse population of addresses and the fact we coordinators didn't initially know any better coupled with the geographical/topological distribution of IP nodes where we couldn't really count on a node being in any specific location within the net. Nodes had to determine their neighbors by discovery and they were routed manually.
I didn't start subnetting until users wanted blocks of IP addresses for specific purposes, like UHF vs VHF gateways, digipeaters, or ARES/RACES. I wrote a paper on it but I don't know how widely it was distributed or how well it was received.
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net
I would be interested too. The project at the moment is to set up essentially an on-demand WISP for disaster response on 3.5 GHz to serve the Monterey Bay area. We will be colocating with one of the significant ISPs here that has a high level site (I do microwave consulting for them, they give me rack space and bandwidth.) We will eventually want a fairly large DHCP pool for use in emergencies. The radios are readily available. I specifically don't want to use NAT'ted private address space because it should be clear that the network is only for traffic that can be handled under the amateur radio rules. There will probably be a trickle of everyday traffic too, remote receivers, linking, etc.
Perhaps we can designate some of the unused /24s for subnetted nodes and leave the existing ones for individual IPs.
WA6NMF
On 2/22/12 6:54 PM, Daniel Curry wrote:
Geoff, Would you mind sending a copy of your paper to the group?
K6DLC
On 2/22/2012 6:47 PM, Geoff Joy wrote:
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF wa6nmf@josephson.com wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can hope!)
I think the reason we started it that way was the sparse population of addresses and the fact we coordinators didn't initially know any better coupled with the geographical/topological distribution of IP nodes where we couldn't really count on a node being in any specific location within the net. Nodes had to determine their neighbors by discovery and they were routed manually.
I didn't start subnetting until users wanted blocks of IP addresses for specific purposes, like UHF vs VHF gateways, digipeaters, or ARES/RACES. I wrote a paper on it but I don't know how widely it was distributed or how well it was received.
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF wa6nmf@josephson.com wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can hope!)
Subnetting is reasonable to do but we still have to assign addresses in those subnets one at a time in order to get DNS entries for them and to enable them in the Internet ingress filter.
The division of the AMPRNet space into the existing blocks of addresses was primarily for administrative convenience, not as a mandated subnetting scheme.
Subnets should probably track routers/gateways; that is, each router/gateway should have a small subnet associated with it. That would help to keep the routing table at a reasonable size. Since routers often serve a specific geographical area, having regional subnets could be a fairly good way to assign addresses.
The hard question is what size region and what size subnet?
The implication is that there will be a router for each region, which is what we've been doing in many places anyway. Perhaps major cities is a reasonable way to divide an area into subnets. But there are also flat networks which need only one router even though they span multiple cities.
Ideas? - Brian
On Wed, Feb 22, 2012 at 19:54, Brian Kantor Brian@ucsd.edu wrote:
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF <
wa6nmf@josephson.com> wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can
hope!)
Subnetting is reasonable to do but we still have to assign addresses in those subnets one at a time in order to get DNS entries for them and to enable them in the Internet ingress filter.
The division of the AMPRNet space into the existing blocks of addresses was primarily for administrative convenience, not as a mandated subnetting scheme.
Subnets should probably track routers/gateways; that is, each router/gateway should have a small subnet associated with it. That would help to keep the routing table at a reasonable size. Since routers often serve a specific geographical area, having regional subnets could be a
fairly
good way to assign addresses.
The hard question is what size region and what size subnet?
The implication is that there will be a router for each region, which is what we've been doing in many places anyway. Perhaps major cities is a reasonable way to divide an area into subnets. But there are also flat networks which need only one router even though they span multiple cities.
Ideas? - Brian
Dynamic DNS to update amprgw's filter.
We only need the filter for ingress to Net44, we could use "Established" and "Related" IPTABLES for egress.
DHCP for mobile users (those moving between access points).
We already have "region" subnets, e.g. 44.x.0.0/17 blocks already allocated to coordinators. That means the inter region routing tables can be relatively small. Within the region subnetting and even sub-subnet can performed based on local need.
I wrote this in late January and you see more comments in the thread starting at http://groups.yahoo.com/group/STARnetDigital/message/209
One of the uses I foresee for STARnet Digital is for it to support "VPNs" for D-STAR Digital Data. Currently, the D-STAR frame addresses one and only one destination. The UR is either set to the gateway for NATing out to the Internet, or it is set to the call of a remote system.
If the UR were to be set to STARnet Digital group, then each frame could be relayed back out to each terminal in the VPN/Group. This still needs to be tested and probably refined.
I think the Net-44 address space could be the unifying point for IP based amateur communications. The NxN text tables being distributed now to IPIP tunnel pockets of activity doesn't scale well and uses a format designed around a specific application. I have been thinking, instead we should build a network around regional routers that each support one 44.x.x.x/16 address space (44.0-255.x.x) -- these could exist in a VPN (maybe LT2P) creating tunnels either to each other or through 1 or more continental/country routers.
In turn, these 256 POP routers, would support up to 256+ local networks (44.x.x.x/24), which in turn could distribute out to progressively smaller and smaller CIDR address spaces.
When AMPRNET was created, the available hardware was either severely limited or financially unreachable for a hobby pursuit. Now a US$40 router (http://routerboard.com/RB750 IPV4, IPV6, VPN, Tunnels, MPLS) can be pressed into service to provide these services ( http://wiki.mikrotik.com/wiki/Manual:License#License_Levels) for any local jump off point to RF (even to a mesh or PTP high speed microwave link). The core routers can be had in the US$350 range (http://routerboard.com/RB1200). There are a number of hams that own or have access to high bandwidth enabled data centers to house core and regional routers.
Additionally, with a little creativity we could build a special DHCP that would examine the D-STAR, AX.25, or ??? frames to assign a Dynamic DNS address to each station (assuming amateur-relay.net as the domain, could be ampr.org):
- In AX.25, if the source address was WF7R and the SSID was 2 then Dynamic DNS records would be created: - wf7r-2 IN A 44.24.73.2 ; Using wf7r-2.amateur-relay.net domain name set IP to 44.24.73.2 - 2.73.24.44.in-addr.arpa. PTR wf7r-2.amateur-relay.net. ; Shows wf7r-2.amateur-relay.net on hostname lookup of IP 44.24.73.2 - On D-STAR, if the mycall address was K7VE and the 8th character (terminal ID) was C, then Dynamic DNS records would be created: - k7ve-c IN A 44.24.88.230 - 230.88.24.44.in-addr.arpa PTR k7ve-c.amateur-relay.net.
Fixed stations and servers likely would have static IPs, but mobile stations, say D-STAR DD units moving from repeater/access point to repeater/access point could release and renew LAN IP addresses using DHCP.
A STARnet Digital server could be modified to include a DHCP lease block for stations in the group/VPN, so mobile D-STAR stations would retain the same DNS entries moving from one repeater/access point to another.
As the owner of the STARnetDigital Yahoo! forum, I invite anyone interested in this topic to reply to the thread there.
------------------------------ John D. Hays K7VE PO Box 1223, Edmonds, WA 98020-1223 http://k7ve.org/blog http://twitter.com/#!/john_hays http://www.facebook.com/john.d.hays
Greetings, Here in Michigan we once had (1990's) a significant number of mobile Hams (truckers and salesman, etc.) that used the RSPF "Radio Shortest Path First" protocol that is included in the JNOS software.
#undef RSPF 1 /* Include Radio Shortest Path First Protocol */
It worked much like what you describe below, except that each mobile was assigned a fixed IP address (from a very specific subnet 44.102.0/24 - the ZERO or MOBILE subnet for Michigan). Each JNOS node in the network 'knew' how to route this subnet for any mobile users as they passed from node to node to node as they drove along. The nodes would 'hand off' a mobile units connection as s/he drove from cell to cell.
Don't ask me the specifics, as I never used this myself nor did I have anything to do with the necessary JNOS configurations to support it. But it worked REALLY well. I know one truck driver with a laptop stayed connected as he drove from Port Huron to Toledo without ever losing his connection (cringes at the thought of texting while driving an 18 wheeler - at night!).
Might be worth investigating rather than re-inventing the wheel...
--- Jay Nugent WB8TKL o Chair, ARRL Michigan Section "Digital Radio Group" (DRG) [www.MI-DRG.org]
On Wed, 22 Feb 2012, K7VE - John wrote:
On Wed, Feb 22, 2012 at 19:54, Brian Kantor Brian@ucsd.edu wrote:
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF <
wa6nmf@josephson.com> wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can
hope!)
Subnetting is reasonable to do but we still have to assign addresses in those subnets one at a time in order to get DNS entries for them and to enable them in the Internet ingress filter.
The division of the AMPRNet space into the existing blocks of addresses was primarily for administrative convenience, not as a mandated subnetting scheme.
Subnets should probably track routers/gateways; that is, each router/gateway should have a small subnet associated with it. That would help to keep the routing table at a reasonable size. Since routers often serve a specific geographical area, having regional subnets could be a
fairly
good way to assign addresses.
The hard question is what size region and what size subnet?
The implication is that there will be a router for each region, which is what we've been doing in many places anyway. Perhaps major cities is a reasonable way to divide an area into subnets. But there are also flat networks which need only one router even though they span multiple cities.
Ideas? - Brian
Dynamic DNS to update amprgw's filter.
We only need the filter for ingress to Net44, we could use "Established" and "Related" IPTABLES for egress.
DHCP for mobile users (those moving between access points).
We already have "region" subnets, e.g. 44.x.0.0/17 blocks already allocated to coordinators. That means the inter region routing tables can be relatively small. Within the region subnetting and even sub-subnet can performed based on local need.
I wrote this in late January and you see more comments in the thread starting at http://groups.yahoo.com/group/STARnetDigital/message/209
One of the uses I foresee for STARnet Digital is for it to support "VPNs" for D-STAR Digital Data. Currently, the D-STAR frame addresses one and only one destination. The UR is either set to the gateway for NATing out to the Internet, or it is set to the call of a remote system.
If the UR were to be set to STARnet Digital group, then each frame could be relayed back out to each terminal in the VPN/Group. This still needs to be tested and probably refined.
I think the Net-44 address space could be the unifying point for IP based amateur communications. The NxN text tables being distributed now to IPIP tunnel pockets of activity doesn't scale well and uses a format designed around a specific application. I have been thinking, instead we should build a network around regional routers that each support one 44.x.x.x/16 address space (44.0-255.x.x) -- these could exist in a VPN (maybe LT2P) creating tunnels either to each other or through 1 or more continental/country routers.
In turn, these 256 POP routers, would support up to 256+ local networks (44.x.x.x/24), which in turn could distribute out to progressively smaller and smaller CIDR address spaces.
When AMPRNET was created, the available hardware was either severely limited or financially unreachable for a hobby pursuit. Now a US$40 router (http://routerboard.com/RB750 IPV4, IPV6, VPN, Tunnels, MPLS) can be pressed into service to provide these services ( http://wiki.mikrotik.com/wiki/Manual:License#License_Levels) for any local jump off point to RF (even to a mesh or PTP high speed microwave link). The core routers can be had in the US$350 range (http://routerboard.com/RB1200). There are a number of hams that own or have access to high bandwidth enabled data centers to house core and regional routers.
Additionally, with a little creativity we could build a special DHCP that would examine the D-STAR, AX.25, or ??? frames to assign a Dynamic DNS address to each station (assuming amateur-relay.net as the domain, could be ampr.org):
- In AX.25, if the source address was WF7R and the SSID was 2 then
Dynamic DNS records would be created: - wf7r-2 IN A 44.24.73.2 ; Using wf7r-2.amateur-relay.net domain name set IP to 44.24.73.2 - 2.73.24.44.in-addr.arpa. PTR wf7r-2.amateur-relay.net. ; Shows wf7r-2.amateur-relay.net on hostname lookup of IP 44.24.73.2
- On D-STAR, if the mycall address was K7VE and the 8th character
(terminal ID) was C, then Dynamic DNS records would be created: - k7ve-c IN A 44.24.88.230 - 230.88.24.44.in-addr.arpa PTR k7ve-c.amateur-relay.net.
Fixed stations and servers likely would have static IPs, but mobile stations, say D-STAR DD units moving from repeater/access point to repeater/access point could release and renew LAN IP addresses using DHCP.
A STARnet Digital server could be modified to include a DHCP lease block for stations in the group/VPN, so mobile D-STAR stations would retain the same DNS entries moving from one repeater/access point to another.
As the owner of the STARnetDigital Yahoo! forum, I invite anyone interested in this topic to reply to the thread there.
John D. Hays K7VE PO Box 1223, Edmonds, WA 98020-1223 http://k7ve.org/blog http://twitter.com/#!/john_hays http://www.facebook.com/john.d.hays
_________________________________________ 44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net
It is my belief that fixed IP addresses are only needed by services (e.g. a web server, irc server, etc.), pretty much all mobile and most base stations do not need fixed addresses.
Being able to address k7ve.ampr.org whether that mobile station is on LAN 1 or LAN 2 is much better than needing to know 44.24.1.10 and having all of the individual routers keeping separate entries for routing to some sort of proxy.
If the demarcation is that anything in 44.0.0.0/8 can address anything else in the 44.0.0.0/8 address space, then sub-netting is a great solution and DNS entries become optional. However, fixed "services" being accessed from outside of 44.0.0.0/8 would still pass through the router (or routers) and they can do a lookup to make sure there is a DNS entry, whether fixed or dynamic.
On Wed, Feb 22, 2012 at 22:23, Jay Nugent jjn@nuge.com wrote:
Greetings, Here in Michigan we once had (1990's) a significant number of mobile Hams (truckers and salesman, etc.) that used the RSPF "Radio Shortest Path First" protocol that is included in the JNOS software.
#undef RSPF 1 /* Include Radio Shortest Path First Protocol */
It worked much like what you describe below, except that each mobile was assigned a fixed IP address (from a very specific subnet 44.102.0/24 - the ZERO or MOBILE subnet for Michigan). Each JNOS node in the network 'knew' how to route this subnet for any mobile users as they passed from node to node to node as they drove along. The nodes would 'hand off' a mobile units connection as s/he drove from cell to cell.
Don't ask me the specifics, as I never used this myself nor did I have anything to do with the necessary JNOS configurations to support it. But it worked REALLY well. I know one truck driver with a laptop stayed connected as he drove from Port Huron to Toledo without ever losing his connection (cringes at the thought of texting while driving an 18 wheeler - at night!).
Might be worth investigating rather than re-inventing the wheel...
--- Jay Nugent WB8TKL o Chair, ARRL Michigan Section "Digital Radio Group" (DRG) [www.MI-DRG.org]On Wed, 22 Feb 2012, K7VE - John wrote:
On Wed, Feb 22, 2012 at 19:54, Brian Kantor Brian@ucsd.edu wrote:
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF <
wa6nmf@josephson.com> wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown
in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can
hope!)
Subnetting is reasonable to do but we still have to assign addresses in those subnets one at a time in order to get DNS entries for them and to enable them in the Internet ingress filter.
The division of the AMPRNet space into the existing blocks of addresses was primarily for administrative convenience, not as a mandated subnetting scheme.
Subnets should probably track routers/gateways; that is, each router/gateway should have a small subnet associated with it. That would help to keep the routing table at a reasonable size. Since routers often serve a specific geographical area, having regional subnets could be a
fairly
good way to assign addresses.
The hard question is what size region and what size subnet?
The implication is that there will be a router for each region, which is what we've been doing in many places anyway. Perhaps major cities is a reasonable way to divide an area into subnets. But there are also flat networks which need only one router even though they span multiple cities.
Ideas? - Brian
Dynamic DNS to update amprgw's filter.
We only need the filter for ingress to Net44, we could use "Established" and "Related" IPTABLES for egress.
DHCP for mobile users (those moving between access points).
We already have "region" subnets, e.g. 44.x.0.0/17 blocks already allocated to coordinators. That means the inter region routing tables can be relatively small. Within the region subnetting and even sub-subnet can performed based on local need.
I wrote this in late January and you see more comments in the thread starting at http://groups.yahoo.com/group/**STARnetDigital/message/209http://groups.yahoo.com/group/STARnetDigital/message/209
One of the uses I foresee for STARnet Digital is for it to support "VPNs" for D-STAR Digital Data. Currently, the D-STAR frame addresses one and only one destination. The UR is either set to the gateway for NATing out to the Internet, or it is set to the call of a remote system.
If the UR were to be set to STARnet Digital group, then each frame could be relayed back out to each terminal in the VPN/Group. This still needs to be tested and probably refined.
I think the Net-44 address space could be the unifying point for IP based amateur communications. The NxN text tables being distributed now to IPIP tunnel pockets of activity doesn't scale well and uses a format designed around a specific application. I have been thinking, instead we should build a network around regional routers that each support one 44.x.x.x/16 address space (44.0-255.x.x) -- these could exist in a VPN (maybe LT2P) creating tunnels either to each other or through 1 or more continental/country routers.
In turn, these 256 POP routers, would support up to 256+ local networks (44.x.x.x/24), which in turn could distribute out to progressively smaller and smaller CIDR address spaces.
When AMPRNET was created, the available hardware was either severely limited or financially unreachable for a hobby pursuit. Now a US$40 router (http://routerboard.com/RB750 IPV4, IPV6, VPN, Tunnels, MPLS) can be pressed into service to provide these services ( http://wiki.mikrotik.com/wiki/**Manual:License#License_Levelshttp://wiki.mikrotik.com/wiki/Manual:License#License_Levels) for any local jump off point to RF (even to a mesh or PTP high speed microwave link). The core routers can be had in the US$350 range ( http://routerboard.com/RB1200**). There are a number of hams that own or have access to high bandwidth enabled data centers to house core and regional routers.
Additionally, with a little creativity we could build a special DHCP that would examine the D-STAR, AX.25, or ??? frames to assign a Dynamic DNS address to each station (assuming amateur-relay.net as the domain, could be ampr.org):
- In AX.25, if the source address was WF7R and the SSID was 2 then
Dynamic DNS records would be created: - wf7r-2 IN A 44.24.73.2 ; Using wf7r-2.amateur-relay.net domain
name set IP to 44.24.73.2 - 2.73.24.44.in-addr.arpa. PTR wf7r-2.amateur-relay.net. ; Shows wf7r-2.amateur-relay.net on hostname lookup of IP 44.24.73.2
- On D-STAR, if the mycall address was K7VE and the 8th character
(terminal ID) was C, then Dynamic DNS records would be created: - k7ve-c IN A 44.24.88.230 - 230.88.24.44.in-addr.arpa PTR k7ve-c.amateur-relay.net.
Fixed stations and servers likely would have static IPs, but mobile stations, say D-STAR DD units moving from repeater/access point to repeater/access point could release and renew LAN IP addresses using DHCP.
A STARnet Digital server could be modified to include a DHCP lease block for stations in the group/VPN, so mobile D-STAR stations would retain the same DNS entries moving from one repeater/access point to another.
As the owner of the STARnetDigital Yahoo! forum, I invite anyone interested in this topic to reply to the thread there.
John D. Hays K7VE PO Box 1223, Edmonds, WA 98020-1223 http://k7ve.org/blog <http://twitter.com/#!/john_**hayshttp://twitter.com/#!/john_hays
<http://www.facebook.com/john.**d.hayshttp://www.facebook.com/john.d.hays
44Net mailing list 44Net@hamradio.ucsd.edu http://hamradio.ucsd.edu/mailman/listinfo/44net
Greetings,
On Wed, 22 Feb 2012, Brian Kantor wrote:
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF wa6nmf@josephson.com wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can hope!)
Subnetting is reasonable to do but we still have to assign addresses in those subnets one at a time in order to get DNS entries for them and to enable them in the Internet ingress filter.
The division of the AMPRNet space into the existing blocks of addresses was primarily for administrative convenience, not as a mandated subnetting scheme.
Subnets should probably track routers/gateways; that is, each router/gateway should have a small subnet associated with it. That would help to keep the routing table at a reasonable size. Since routers often serve a specific geographical area, having regional subnets could be a fairly good way to assign addresses.
The hard question is what size region and what size subnet?
The implication is that there will be a router for each region, which is what we've been doing in many places anyway. Perhaps major cities is a reasonable way to divide an area into subnets. But there are also flat networks which need only one router even though they span multiple cities.
Ideas?
Michigan was assigned 44.102/16 and shortly after that we had a group of interested parties get together and define a state "plan". It went something like this...
Michigan has 83 counties, however the Upper Penninsula had already been claimed by Wisconsin as part of their coordination region. We then determined how many licensed Hams there were in each county - this gave us some idea just how much or little AMPRnet activity to expect in each county. We then assigned anywhere from TWO to EIGHT /24 subnets to each county based on their Ham populations.
For example: Wayne county (LOTS of Hams!) received 8 subnets 44.102.48/24 through 44.102.55/24
Mason county (very few Hams) recieved 2 subnets 44.102.146/24 and 44.102.147/24
This assured a better "distribution" of the IP address space.
Jeff King WB8WKA was the Michigan IP Address coordinator at that time. Later I took over (Jay Nugent WB8TKL) the reines when Jeff moved on to other interests.
The following is a JPG image of the map I wrote up back in 1992. The numbers listed on each county defines the range of the 3rd octets for that county's subnets:
http://www.mi-drg.org/AMPR-subnets-by-county.jpg
This plan continues to work to this day! We have had situations where one Gateway/Hamgate supported routes for multiple counties. And we have had multiple Gateways/Hamgates within one county with each agreeing to route one or more subnets - but never do BOTH Hamgates route the SAME subnet. Users could get an IP address for EACH gateway and choose which would be their default route.
We have assigned entire subnets and or blocks/ranges of addresses to groups or individuals for their own experimentation.
We also promote that each county operate on their own frequency to avoid the "Hidden Transmitter" problems. We also reccommend that subnets exist on their own frequencies as well.
There are currently 277 route entires in todays ENCAP.TXT route table. With 49 of those being Michigan /24 (a couple are /23 or /22) subnets used in our counties. We make up about 17% of the total ENCAP route table. Perhaps that is a sign as to our level of activity? I can only hope...
Enjoy! --- Jay Nugent WB8TKL o Chair, ARRL Michigan Section "Digital Radio Group" (DRG) [ www.MI-DRG.org ] o Michigan AMPRnet IP Address Coordinator [ http://www.mi-drg.org/ip-addr-form.html ]
() ascii ribbon campaign in /\ support of plain text e-mail +------------------------------------------------------------------------+ | Jay Nugent jjn@nuge.com (734)484-5105 (734)649-0850/Cell | | Nugent Telecommunications [www.nuge.com] | | Internet Consulting/Linux SysAdmin/Engineering & Design/ISP Reseller | | ISP Monitoring [www.ispmonitor.org] ISP & Modem Performance Monitoring | | Web-Pegasus [www.webpegasus.com] Web Hosting/DNS Hosting/Shell Accts| +------------------------------------------------------------------------+ 23:01:01 up 171 days, 5:39, 3 users, load average: 0.00, 0.02, 0.00
On 2/22/12 7:54 PM, Brian Kantor wrote:
Subnetting is reasonable to do but we still have to assign addresses in those subnets one at a time in order to get DNS entries for them and to enable them in the Internet ingress filter.
I understand the intent ... but as Jay WB8TKL pointed out there are lots of assignments that have no DNS (or listing in amprhosts).
The division of the AMPRNet space into the existing blocks of addresses was primarily for administrative convenience, not as a mandated subnetting scheme.
Subnets should probably track routers/gateways; that is, each router/gateway should have a small subnet associated with it. That would help to keep the routing table at a reasonable size. Since routers often serve a specific geographical area, having regional subnets could be a fairly good way to assign addresses.
The hard question is what size region and what size subnet?
At present (at least per Daniel Curry, new coordinator for my region) coordinators seem to have been instructed to assign individual addresses, not subnets, in order to capture per-host domain information. This causes a problem in that when one goes to register a gateway, the range of addresses doesn't necessarily fall into a normal power of 2 as would be expected for a subnet. I don't think there is a lot of contention for this address space, so you could throw a dart and say that anyone running a gateway should be assigned a /24 or /26 or such. But then, there's the challenge of how to get data for DNS and ingress filtering as you mention.
Perhaps the gateway robot can be modified to expect entries for subnets and hosts within that subnet, including the desired DNS entry for each host. Then the operator of the gateway could be responsible for assigning local addresses, maintaining the DNS entries for each, and sending those updates via the robot.
73 de WA6NMF
On 23 Feb 2012, at 03:54, Brian Kantor wrote:
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF wa6nmf@josephson.com wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can hope!)
Subnetting is reasonable to do but we still have to assign addresses in those subnets one at a time in order to get DNS entries for them and to enable them in the Internet ingress filter.
The division of the AMPRNet space into the existing blocks of addresses was primarily for administrative convenience, not as a mandated subnetting scheme.
The hard question is what size region and what size subnet?
1. That is indeed a hard question. And I don't think one size will fit all no matter how much effort goes into it. My own preference is logical areas with 'room to grow'. However as usage of IP has died a death here.
2. I'd be interested in seeing the paper as well
and 3. Going back to startup scripts.
With a lot of reading old posts and figuring things out. I ended up with the following (Ubuntu 10.04 LTS)
root@mini-itx:~# cat /etc/iproute2/rt_tables # # reserved values # 255 local 254 main 253 default 0 unspec # # local # #1 inr.ruhep 200 44
/etc/network/interfaces .. auto iface tunl1 inet static address 44.155.6.12 netmask 255.255.255.0 network 44.155.6.0 broadcast 44.155.6.255 pre-up /sbin/ip tunnel add tunl1 mode ipip remote 169.228.66.251 local 10.100.12.254 post-down /sbin/ip tunnel del tunl1 pre-down /sbin/ip route del default via 169.228.66.251 onlink table 44 pre-down /sbin/ip route del 169.228.66.251 dev eth0 table 44 pre-down /sbin/ip rule del to 44.0.0.0/8 table 44 pre-down /sbin/ip rule del from 44.0.0.0/8 table 44 pre-down /sbin/ip rule del to 44.155.6.12/24 table main up /sbin/ip rule add to 44.155.6.12/24 table main priority 1 up /sbin/ip rule add from 44.0.0.0/8 table 44 priority 44 up /sbin/ip rule add to 44.0.0.0/8 table 44 priority 45 up /sbin/ip route add 169.228.66.251 dev eth0 table 44 up /sbin/ip route add default dev tunl1 via 169.228.66.251 onlink table 44
I'm using the most excellent rip44d
I found (by trial, error, and confusion) that if I had another interface with the 44.155.6.12 IP. rip44d wouldn't work. I'm using 44.155.6.12 as its one of the only 2 IPs that I have that is already in the DNS.
Cheers John EI7IG
The first few requests I received for 44 net addresses came from MSYS BBS operators wanting an IP address. In every case, it was a very short time before they asked for another address or two. Eventually, when I got requests regardless from who or why, I just assigned a block of 4 addresses, then 8. The addresses naturally fell on boundaries that could be segregated by a netmask. They became de facto subnets. I made it clear that was not the purpose but merely convenient for both of us. The node operators of Netrom nodes were continually installing TheNET X1J and instituting subnets for IP routing. Basically, the hardware or software was driving the implementation of the network. Not the other way around.
As I understand it, the intent was to develop a network with enough sophistication that mobile stations (IP addresses) could remain connected because the network would dynamically discover and track/route without having the operator consciously change IP address or any other identifier. RSPF is one effort toward that goal.
__Reid, WB7CJO__
Reid Fletcher Chief Engineer, KUWR Wyoming Public Radio - University of Wyoming
On 2/22/2012 7:47 PM, Geoff Joy wrote:
On Wed, 22 Feb 2012 15:19:31 -0800, David Josephson WA6NMF wa6nmf@josephson.com wrote:
I am puzzled that we want to assign 44-net addresses one by one as shown in amprhosts rather than as subnets. Perhaps there is a historical reason for that. The routing table could get to be very large (we can hope!)
I think the reason we started it that way was the sparse population of addresses and the fact we coordinators didn't initially know any better coupled with the geographical/topological distribution of IP nodes where we couldn't really count on a node being in any specific location within the net. Nodes had to determine their neighbors by discovery and they were routed manually.
I didn't start subnetting until users wanted blocks of IP addresses for specific purposes, like UHF vs VHF gateways, digipeaters, or ARES/RACES. I wrote a paper on it but I don't know how widely it was distributed or how well it was received.
-
Brian Kantor -
Daniel Curry -
David Josephson -
David Josephson WA6NMF
-
Donald Jacob -
Geoff Joy -
Jay Nugent -
John Ronan -
K7VE - John -
Marius Petrescu -
Reid Fletcher -
WA6NMF