IPv6 Subnetting

[MoonRise]

So a 288000 in² home would have that times 23 huh? Hmmmm what can one develop in the way of super computer. Hopefully no Skynet!


For us Star Trekies, I wonder what the Borg use? Has that ever been stated?

[SteveRiley]

It isn't uncommon to block ICMP. Check the man pages for those tools... see if you can force either of the traceroutes to connect to destination port 53/udp. That's what a DNS server listens on.

[GreyGeek]

The package ndisc6 includes a number of IPv6 diagnostic tools. Try running tcptraceroute6 against the OpenDNS servers.

Nice set of tools! Thanks for the heads-up! :cool:

The tool traceroute6 times out after 30 hops...

9:~$ traceroute6 2620:0:ccc::2
traceroute to 2620:0:ccc::2 (2620:0:ccc::2) from 2001:4978:f:580::2, 30 hops max, 24 byte packets
1 gw-1409.chi-02.us.sixxs.net (2001:4978:f:580::1) 27.626 ms 27.715 ms 28.21 ms
2 unassigned.v6.your.org (2001:4978:1:400::53:1) 28.316 ms 29.814 ms 27.818 ms
3 * sixxs.ge-0.0.0-30.core1.chi.bb6.your.org (2001:4978:1:400::ffff) 142.557 ms 29.225 ms
4 ibx-chi.ip6.tiscali.net (2001:504:0:4::3257:1) 29.24 ms 30.046 ms 31.515 ms
5 xe-5-1-0.was10.ip6.tinet.net (2001:668:0:2::1:f2) 57.52 ms * 57.946 ms
6 opendns-gw.ip6.tinet.net (2001:668:0:3::8000:18d2) 61.802 ms * 56.533 ms
7 * * *
8 * * *
...

Tcptraceroute6 appears to give the same listing.

:~$ tcptraceroute6 2620:0:ccc::2
traceroute to 2620:0:ccc::2 (2620:0:ccc::2) from 2001:4978:f:580::2, port 80, from port 49739, 30 hops max, 60 bytes packets
1 * gw-1409.chi-02.us.sixxs.net (2001:4978:f:580::1) 29.717 ms 29.501 ms
2 unassigned.v6.your.org (2001:4978:1:400::53:1) 30.154 ms 29.938 ms 31.550 ms
3 sixxs.ge-0.0.0-30.core1.chi.bb6.your.org (2001:4978:1:400::ffff) 29.965 ms 32.059 ms 30.863 ms
4 ibx-chi.ip6.tiscali.net (2001:504:0:4::3257:1) 33.447 ms 31.769 ms 36.159 ms
5 xe-5-1-0.was10.ip6.tinet.net (2001:668:0:2::1:f2) 58.810 ms 57.230 ms 60.254 ms
6 opendns-gw.ip6.tinet.net (2001:668:0:3::8000:18d2) 55.090 ms 57.779 ms 77.314 ms
7 * * *
...

All of the sites listed relate to SixXS. I was surprised to see the 6th site, an "opendns" site. it must not work, because after than the trace times out.

[SteveRiley]

The package ndisc6 includes a number of IPv6 diagnostic tools. Try running tcptraceroute6 against the OpenDNS servers.

[GreyGeek]

mmm... That pretty much looks like mine.

Steve pretty much explained why you get a 10/10 ... it's because your ISP supports IPv6 natively. TimeWarner gives an AAAA record when I "host -t AAAA timewarnercable.com" but in reading between the lines of that "1%" support figure, it seems that they are the "business" and "premium" class users. The peasants will get IPv6 "real soon now".

My /etc/NetworkManager/system-connections/GreyGeek config file (your's will be the name you assigned to your AP) contains:

[ipv6]
method=auto
dns=2620:0:ccc::2;2620:0:ccd::2;
dns-search=opendns.com;
ignore-auto-dns=true

I added the opendns.com DNS addresses via the IPv6 tab of the NetworkManager Connections Editor. But, the opendns.com dns numbers are ignored. I've experimented with commenting out the "method" and "ignore..." lines in the four possible combinations but it doesn't help. I always get 9/10.

Ping6ing you IPv6, 2001:0:53aa:64c:28e9:18d8:d1c4:ceb7, returns nothing. I can't find your or my IPv6 address using http://ip-lookup.net/.

[Jonas]

Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
0.0.0.0         192.168.0.1     0.0.0.0         UG    0      0        0 eth0
169.254.0.0     0.0.0.0         255.255.0.0     U     1000   0        0 eth0
192.168.0.0     0.0.0.0         255.255.255.0   U     1      0        0 eth0

[GreyGeek]

Yup, that is what I was looking for, but it doesn't show me anything that would explain why you get a 10/10. Mine looks pretty much the same except it is with aicuu and SixXS.

What does your route table look like?
sudo route -n

[Jonas]

@ GG

This is output; (what are you looking for? )

jonas@kubuntu:~$ ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 14:da:e9:00:14:93 brd ff:ff:ff:ff:ff:ff
inet 192.168.0.181/24 brd 192.168.0.255 scope global eth0
inet6 fe80::16da:e9ff:fe00:1493/64 scope link
valid_lft forever preferred_lft forever
4: teredo: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1280 qdisc pfifo_fast state UNKNOWN qlen 500
link/none
inet6 2001:0:53aa:64c:c47:3ccc:d1c4:ceb7/32 scope global
valid_lft forever preferred_lft forever
inet6 fe80::ffff:ffff:ffff/64 scope link
valid_lft forever preferred_lft forever
jonas@kubuntu:~$

[Jonas]

GG - sorry missunderstood your question. installed thru repo. On mobile now away from my box. Will check when I get home in a couple of hours.

b.r
Jonas

[GreyGeek]

....
Is Teredo local or like a proxy ?
...

Neither.

Did you install miredo from the repository? It shows itself with a dev listing of "teredo" when you use
ip addr

Miredo creates a "tunnel" between your PC and a server. When I was investigating it the server was in Germany. There is probably more than one server out there.

The tunnel created using SixXS as the serverr, and aicuu (in the repository) as a driver gives me an IPv6 address of
2001:4978:f:580::2
and at the other end of the tunnel is the address 2001:4978:f:580::1

:~$ ping6 2001:4978:f:580::2
PING 2001:4978:f:580::2(2001:4978:f:580::2) 56 data bytes
64 bytes from 2001:4978:f:580::2: icmp_seq=1 ttl=64 time=0.043 ms
64 bytes from 2001:4978:f:580::2: icmp_seq=2 ttl=64 time=0.038 ms
64 bytes from 2001:4978:f:580::2: icmp_seq=3 ttl=64 time=0.054 ms
64 bytes from 2001:4978:f:580::2: icmp_seq=4 ttl=64 time=0.048 ms
64 bytes from 2001:4978:f:580::2: icmp_seq=5 ttl=64 time=0.049 ms
^C
....
PING 2001:4978:f:580::1(2001:4978:f:580::1) 56 data bytes
64 bytes from 2001:4978:f:580::1: icmp_seq=1 ttl=64 time=28.7 ms
64 bytes from 2001:4978:f:580::1: icmp_seq=2 ttl=64 time=209 ms
64 bytes from 2001:4978:f:580::1: icmp_seq=3 ttl=64 time=30.0 ms
64 bytes from 2001:4978:f:580::1: icmp_seq=4 ttl=64 time=28.2 ms
^C
--- 2001:4978:f:580::1 ping statistics ---

:~$ ip r s t all
default via 192.168.1.1 dev wlan0 proto static <--- IPv4 gateway
169.254.0.0/16 dev wlan0 scope link metric 1000
192.168.1.0/24 dev wlan0 proto kernel scope link src 192.168.1.100 metric 2
broadcast 127.0.0.0 dev lo table local proto kernel scope link src 127.0.0.1
local 127.0.0.0/8 dev lo table local proto kernel scope host src 127.0.0.1
local 127.0.0.1 dev lo table local proto kernel scope host src 127.0.0.1
broadcast 127.255.255.255 dev lo table local proto kernel scope link src 127.0.0.1
broadcast 192.168.1.0 dev wlan0 table local proto kernel scope link src 192.168.1.100
local 192.168.1.100 dev wlan0 table local proto kernel scope host src 192.168.1.100
broadcast 192.168.1.255 dev wlan0 table local proto kernel scope link src 192.168.1.100
::/96 via :: dev sit0 metric 256
2001:4978:f:580::/64 dev sixxs proto kernel metric 256
fe80::/64 dev wlan0 proto kernel metric 256
fe80::/64 dev sixxs proto kernel metric 256
default via 2001:4978:f:580::1 dev sixxs metric 1024 <--- IPv6 connection.

[Jonas]

Most likely.

If your IPv6 address is in the 2000::/3 network (that's 2000:: to 3FFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF), then you have a publicly routable globally unique address (this constitutes one eigth of the entire IPv6 address space). Within this range are some exceptions:

• 2001::/29 - 2001:1f8::/29 is set aside for various sub-top-level aggregation (which includes the Teredo range 2001://32)
• 2001:db8://32 is for documentation purposes and is (supposed to be) explicitly non-routable
• 2002::/16 is for 6to4 (two IPv6 hosts communicating over an IPv4-only network)

Oh, and don't forget. In IPv6-land, there's no place like ::1

Hmm .. not sure I fully understand it all, networking (digital nor social lol ) have never been my strong side. I guess I'll understand later on, when trouble start cropping up

Wow, I've been busting my bums trying to get 10/10.

What does "ip addr" show on your box?

If this is what you meant;

Din publika IPv6 adress ser ut att vara 2001:0:53aa:64c:28e9:18d8:d1c4:ceb7
Din IPv6-tjänst ser ut att vara: Teredo

Is Teredo local or like a proxy ?

b.r

Jonas

PS/edit; If anyone (like I had) have trouble to get IPv6 showing up at all - it may be because it's disabled in Firefox > switch value; true to false in about:config > network.dns.disableIPv6 string

[GreyGeek]

....
Edit; testing the links in previous posts give me IPv4 - 10/10 and IPv6 - 10/10 with Teredo, and DNS/ISP is IPv6 ready.
...

Wow, I've been busting my bums trying to get 10/10.

What does "ip addr" show on your box?

[SteveRiley]

Does he refer to "pure" IPv6 ?

Most likely.

If your IPv6 address is in the 2000::/3 network (that's 2000:: to 3FFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF), then you have a publicly routable globally unique address (this constitutes one eigth of the entire IPv6 address space). Within this range are some exceptions:

• 2001::/29 - 2001:1f8::/29 is set aside for various sub-top-level aggregation (which includes the Teredo range 2001://32)
• 2001:db8://32 is for documentation purposes and is (supposed to be) explicitly non-routable
• 2002::/16 is for 6to4 (two IPv6 hosts communicating over an IPv4-only network)

Oh, and don't forget. In IPv6-land, there's no place like ::1

[Jonas]

Thanks for sharing this GG! I wasn't aware that my ISP (Bahnhof) provided IPv6, according to wikipedia .SE - (google translated) "Internet is still dominated by IPv4, and only a few percent of the Internet computers have IPv6 addresses. In Sweden there is IPv6 support from, among others Bredband2, IP-Only, SUNET, Phonera, Loopia and Bahnhof and Hi3G_Access_AB (3). In the U.S. it has been decided that all federal agencies must have some support for IPv6 by 2008." - https://sv.wikipedia.org/wiki/IPv6

However I'm puzzled, reading this article Bahnhof claim they don't provide IPv6 to end users.

http://www.zdnet.com/blog/norse-code...-the-world/103

"Johnny Aspman, chief executive officer of Bahnhof says it has IPv6 in trial mode today, but it can’t deliver IPv6 to end users until the 80-plus owners behind the 150 networks that make up Sweden’s open access last mile network swap out infrastructure to support IPv6 or dual stack.
That means networks might offer IPv6 on a business to business basis, but not a single operator in Sweden offers IPv6 to end-users commercially.

“Everybody’s talking about it, but it takes time,” says Bahnhof’s Aspman."

Does he refer to "pure" IPv6 ?


Edit; testing the links in previous posts give me IPv4 - 10/10 and IPv6 - 10/10 with Teredo, and DNS/ISP is IPv6 ready.
b.r
Jonas

[SteveRiley]

Jees, guys. I keep reading Jerry's posts hoping I can learn geek that way. Doesn't seem to help, it's still geek to me. So is the ipv6 a way to improve the usability of present bandwidth? Is this something I need to start to configure for?

IPv6 is the replacement network protocol for the currently-pervasive IPv4, designed in the early 1970s. IPv4 uses a 32-bit addressing space. 2^32 was a huge number in 1970; it isn't so big anymore. All available address blocks have been assigned to regional numbering authorities. While most numbering authorities still have a decent number of addresses to hand out, APNIC (for Asia) is close to running out.

IPv6 uses a 128-bit addressing space. 2^128 is an impossibly huge number. IPv6 fixes a number of deficiencies not envisioned during the design of IPv4. Mostly, these solve problems related to device mobility, something not imagined 40 years ago. IPv6's mechanisms for automatic address assignment, stateless configuration, router/neighbor discovery, and origin privacy are vastly improved. Routing is simplified, and asymmetric ("triangular") routing disappears. But IPv6 also introduces some new levels of complexity, mostly related to IPv4-IPv6 coexistence. And because it's still so new for so many people, IPv6 networks are likely to be riddled with configuration vulnerabilities for a while, as people learn.

IPv6 won't increase or decrease effective bandwidth. IPv6 has no mechanism for full broadcasting, instead relying on multicasting, so in theory IPv6 might reduce overall utilization a little bit.