1 Oct
2016
1 Oct
'16
5:39 p.m.
The good thing about IPv6 is that due to the large
address space it's
possible to make it mandatory to encode the callsign in the last 64 bits
of the address. This eliminates the problem that we have in IPv4 of
trying to trace back which callsign is behind an IP.
I think it basically is a good idea, but it needs to be more flexible.
We would like to be able to derive the callsign from the IP, but there should
be no 1:1 mapping between callsign and IP, because that would mean only a single
system can be online for each callsign.
The standard should leave some bits (at least 8) available for use as "SSID" as
we had in the packet days (callsign-1 callsign-2 etc), maybe also with some
encoding of alphanumeric values.
Rob
1 Oct
1 Oct
5:45 p.m.
New subject: IPv6
The most popular allocations from ISPs I have seen are either a /48 or a /56. Since /56 is
the smallest I have found, maybe we can also use these 8-bits (bit 56 to bit 64) for a
special meaning, such as the area (K4, SV2, etc.), or the country (K, SV, etc.). That way
each person can have a ham subnet, and then maybe use the first bits of the remaining 64
to encode more information. This however breaks SLAAC, but assuming these will be
typically servers, you can configure them statically. Now about the users I am not quite
sure how it will work without resorting to solutions like stateful DHCPv6.
Antonios
On 01 Oct 2016, at 13:39, Rob Janssen
<pe1chl(a)amsat.org> wrote:
(Please trim inclusions from previous messages)
_______________________________________________
The good thing about IPv6 is that due to the
large address space it's
possible to make it mandatory to encode the callsign in the last 64 bits
of the address. This eliminates the problem that we have in IPv4 of
trying to trace back which callsign is behind an IP.
I think it basically is a good idea, but it needs to be more flexible.
We would like to be able to derive the callsign from the IP, but there should
be no 1:1 mapping between callsign and IP, because that would mean only a single
system can be online for each callsign.
The standard should leave some bits (at least 8) available for use as "SSID"
as
we had in the packet days (callsign-1 callsign-2 etc), maybe also with some
encoding of alphanumeric values.
Rob
_________________________________________
44Net mailing list
44Net(a)hamradio.ucsd.edu
http://hamradio.ucsd.edu/mailman/listinfo/44net 
7:14 p.m.
New subject: IPv6
On 2016-10-01 13:45, DaKnOb wrote:
(Please trim inclusions from previous messages)
_______________________________________________
The most popular allocations from ISPs I have seen are either a /48 or a /56. Since /56
is the smallest I have found, maybe we can also use these 8-bits (bit 56 to bit 64)
You are certainly an optimist. Even if general recommendations for ISPs
are to allocate at least a /56 block per user, there are lots of them
actually allocating dynamically only a /64 block.
Even so, an encoding scheme would work...
Fortunately, there are ISPs like Hurricane Electric which offer free
6in4 tunnels for free, with /64 + /48 subnets.
Marius, YO2LOJ
7:53 p.m.
New subject: IPv6
Hmm.. Do you have any examples of such ISPs? I’ve only seen them give out a /64 to the
PPPoE / WAN interface and then also have a routable /56 or /48 which is almost always
available using DHCPv6-PD. The /64 is only used for communication with the ISP (can also
be a /126) while everything else if up to the router to manage.
On 01 Oct 2016, at 15:14, Marius Petrescu
<marius(a)yo2loj.ro> wrote:
You are certainly an optimist. Even if general recommendations for ISPs
are to allocate at least a /56 block per user, there are lots of them
actually allocating dynamically only a /64 block.
10:21 p.m.
New subject: IPv6
RDS-RCS, one of the biggest ISPs in Romania does this. The others don't
offer IPv6...
On 2016-10-01 15:53, DaKnOb wrote:
(Please trim inclusions from previous messages)
_______________________________________________
Hmm.. Do you have any examples of such ISPs? I’ve only seen them give out a /64 to the
PPPoE / WAN interface and then also have a routable /56 or /48 which is almost always
available using DHCPv6-PD. The /64 is only used for communication with the ISP (can also
be a /126) while everything else if up to the router to manage.
On 01 Oct 2016, at 15:14, Marius Petrescu
<marius(a)yo2loj.ro> wrote:
You are certainly an optimist. Even if general recommendations for ISPs
are to allocate at least a /56 block per user, there are lots of them
actually allocating dynamically only a /64 block.
_________________________________________
44Net mailing list
44Net(a)hamradio.ucsd.edu
http://hamradio.ucsd.edu/mailman/listinfo/44net 
5:55 p.m.
New subject: IPv6
El 01/10/16 a las 12:39, Rob Janssen escribió:
(Please trim inclusions from previous messages)
_______________________________________________
Good point, Rob.
I didn't make it clear enough in my previous message, but by "callsign"
I mean "callsign + SSID". What is a valid callsign + SSID depends on
which system to encode callsigns in IPv6's you use.
The system I'm using:
https://github.com/darconeous/ham-addr/blob/master/ham-addr.txt.md
allows the following:
* A 8-character callsign can be encoded into an EUI-48. This can be used
as a MAC for an ethernet device and SLAAC will derive the IPv6
accordingly. With a 6 character base callsign such as EA4GPZ, you can
use EA4GPZ-?, where ? is any character from A to Z or 0 to 9, so you get
36 possible "callsigns + SSID". Actually, you can also encode a 9
character callsign as long as the last character is H, P, X or 5, so you
can get 36*5 = 180 different "callsigns + SSID" using this ugly hack.
* An 11-character callsign can be encoded into an EUI-64. This can be
used directly as the last 64bits of the IP address. With 6 character
base callsign such as EA4GPZ, you can use EA4GPZ-????, so you get 36^4
possible "callsigns + SSID", which I think it's already much more than
anyone will neeed. Also, you can use compound callsigns such as
MW/EA4GPZ/P, should anyone have the need for this.
Other systems allow for different possibilities, but the idea is to
always have some sort of SSID. One IP per callsign is not acceptable. I
already have a router, a server, 2 ubiquiti devices and my laptop all
running under my callsing, using different SSIDs.
73,
Dani.
The good thing about IPv6 is that due to the
large address space it's
possible to make it mandatory to encode the callsign in the last 64 bits
of the address. This eliminates the problem that we have in IPv4 of
trying to trace back which callsign is behind an IP.
I think it basically is a good idea, but it needs to be more flexible.
We would like to be able to derive the callsign from the IP, but there
should
be no 1:1 mapping between callsign and IP, because that would mean only
a single
system can be online for each callsign.
The standard should leave some bits (at least 8) available for use as
"SSID" as
we had in the packet days (callsign-1 callsign-2 etc), maybe also with some
encoding of alphanumeric values.
3008
days inactive
3008
days old
5 comments
4 participants
participants (4)
-
DaKnOb -
Dani EA4GPZ -
Marius Petrescu -
Rob Janssen