IPv6 Subnetting

[GreyGeek]

While experimenting with IPv6 on Precise I found an interesting web page explaining how the 128 binary bits that make up an IPv6 IP address are divided up between the 48 bits which define the Internet address, the 16 bits which are used for the subnet, and the remaining 64 bits which are used for devices that are used on a particular subnet:

http://www.exabyte.net/lambert/subnet/ipv6_subnet_masking.htm

The current IPv4 IP address has 32 bits, which allows for 4,294,967,296 connections.

The 48 bits reserved for Internet addresses in IPv6 allows for 281,474,976,710,656 connections. If there are 7 billion people on this planet those 48 bits will allow EVERYONE on this planet plus EVERYONE on 42,209 other planets as populated as this one to have their own IPv6 address!

But, just like IPv4, some patterns of the 48 Internet address bits are reserved for specific uses.
https://wiki.ubuntu.com/IPv6

Special IPv6 address rangesAddress ranges are specified using a prefix. This is similar to the IPv4 concept of a network prefix and netmask, but in IPv6 we always use CIDR notation to specify the number of bits at the start of the mask that are 1. For example, the address range fe80::/10 includes all addresses that are the same as the address fe80:: in the first 10 bits, i.e. all beginning with the bit pattern 1111 1110 10. One hex digit corresponds to four bits: 1111 is f, 1110 is e. 10 is the most significant two bits of 1000 which is 8 in hex, but since this length is not a multiple of 4, the range also includes addresses with this digit set to 9 (1001), a (1010) and b (1011). Obviously a network prefix of length 128 specifies an individual address.


IPv6 address

Prefix length (bits)

Description

Notes



::

128

unspecified

Used for default route and router solicitations. cf. 0.0.0.0 in IPv4



::1

128

loopback address

cf. 127.0.0.1 in IPv4



::ffff:a.b.c.d

96

IPv4 mapped IPv6 address

The lower 32 bits are the IPv4 address. Used in socket API's to represent IPv4 hosts.



fe80::

10

link-local

Unroutable autoconfigured addresses used on a LAN, e.g. for DHCPv6



fc00::

7

unique local

Addresses used only within an autonomous system, unroutable globally. Cf. RFC 1918 addresses such as used in NAT.



ff00::

8

multicast



2000::

3

global unicast

All global unicast addresses currently begin with 2.




Some IPv6 address prefixes have been deprecated, and should no longer be used.


deprecated IPv6 address

Prefix length (bits)

Description

Notes



3ffe::

16

6bone prefix

Used 1998-2006.



::a.b.c.d

96

embedded IPv4

96 zero bits followed by 32 IPv4 bits. Also called “IPv4 compatible IPv6 address”. Replaced by mapped addresses.



fec0::

10

site-local

Replaced by Unique Local Addresses




Some address prefixes are special use:


special IPv6 address

Prefix length (bits)

Description

Notes



2001:db8::

32

documentation examples

Not to be routed.



2001:0::

32

Teredo tunnels

the remaining bits come from a Teredo server and the client NAT device.



2002::

16

6to4 tunnels

the next 32 bits are the client IPv4 address

[SteveRiley]

Jerry: 2^128 addresses would allow us to have how many addresses per square inch of the earth's surface, including oceans? ;)

[GreyGeek]

Jerry: 2^128 addresses would allow us to have how many addresses per square inch of the earth's surface, including oceans? ;)

Well, now that you've asked! ... ;D

The surface of the earth is 7.9 x 10^17 square inches, according to Wolfram Alpha.

So, each square inch on this planet could have 4.3 X 10^20 IPv6 addresses all their own.

If we limited each square inch to the number of device addresses allowed to each IP address then each square inch of the Earth's surface would have 23 IPv6 device addresses to play with.

I don't think we are going to run out of IPv6 address spaces anytime soon unless we develop intelligent nanobots that began asking for connections. ;D

[SteveRiley]

I figured you'd find this to be a fun exercise :)

[SteveRiley]

fec0:: 10 site-local Replaced by Unique Local Addresses
Some poorly-written stacks still try to create site-local addresses, or attempt to locate DNS servers with the fec0: prefix. An easy way to block all site-local traffic is with a rule that discards everything: block the entire range fec0::ff. I think that's pretty funny ;D

(Pronouce aloud, substitute an "Oh" for the "zero," don't say "colon colon"...)

[GreyGeek]

I figured you'd find this to be a fun exercise :)

It sure does give one a sense of how BIG the domain and device spaces are in IPv6.

[NickStone]

It sure does give one a sense of how BIG the domain and device spaces are in IPv6.

...and my internet connection is still using IPv4

[GreyGeek]

...and my internet connection is still using IPv4

So, when you ping6 your ISP domain name you get "unknown host"?

[NickStone]

So, when you ping6 your ISP domain name you get "unknown host"?

I hadn't tried it before you mentioned it, but have now and the answer is yes. But then I am only using a mobile broadband "dongle" to connect to internet. (T-mobile is what I'm using.

[GreyGeek]

T-Mobil has activated IPv6 across their entire US network.
http://support.t-mobile.com/thread/26025


As of right now, the only devices that support IPv6 over GSM and UMTS are the Samsung Nexus S for T-Mobile USA and the international unlocked HSPA+ Samsung Galaxy Nexus. The deployment of IPv6 is also critical to T-Mobile’s upcoming launch of LTE next year (http://www.extremetech.com/electronics/119703-t-mobile-announces-lte-prepares-to-take-the-us-wireless-market-by-storm). LTE uses IPv6 natively across the network, so all LTE (http://www.extremetech.com/mobile/110711-what-is-lte) devices are IPv6 capable, too. T-Mobile’s HSPA+/LTE devices will be IPv6 capable on all types of supported networks, rather than just LTE and WiFi (which is how Verizon and Sprint LTE devices will work).

It would probably be a minor miracle if you could get Kubuntu to connect, but it may be worth a try. Meanwhile, you could install miredo from the repository and use it to access IPv6 websites.

The T-Mobil setup process is explained here if you are a T-Mobile phone user and are using one of the phones mentioned above.:
https://sites.google.com/site/tmoipv6/lg-mytouch



Known Bugs


Peer-to-peer applications like Skype and Tango do not work with IPv6. If you use these services, please contacts Skype (https://support.skype.com/en-us/)and Tango (http://support.tango.me/home)to request support of IPv6. Here is a list applications (https://docs.google.com/spreadsheet/ccc?key=0AnVbRg3DotzFdGVwZWlWeG5wXzVMcG5qczZEZloxW Gc) that have been tested on the T-Mobile USA IPv6 network
The IPv6 service only works in T-Mobile USA coverage (http://coverage.t-mobile.com/), it does not work in domestic or international roaming at this time.
Tethering is not yet an IPv6 feature in Android (special builds work (http://dan.drown.org/android/clat/))
A-GPS feature in the Nexus phones does not work with IPv6 correctly, this will make locking a GPS signal take longer.
Have another bug to report? Send it to the Google Group (http://groups.google.com/group/tmoipv6beta)




You can test your setup here:
http://ip6.me/

http://test-ipv6.com/

P.S. --
http://support.t-mobile.com/message/147760#147760


If you apply the Android 4.0 ICS update to the Samsung Galaxy S II, you can now connect to the T-Mobile USA network using IPv6!

[NickStone]

Thanks for the info but it won't work for me as I'm using T-Mobile UK. Just looking on their website for IPv6 rollout but so far haven't found anything yet.

[capt-zero]

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?

[GreyGeek]

... So is the ipv6 a way to improve the usability of present bandwidth? Is this something I need to start to configure for?

The IPv4 address space IS exhausted. New webpages can only appear in volume if they are given IPv6 addresses. . 80% of Internet traffic is because of 20 websites, and they have converted to IPv6. They only reason why you can still reach them is because they still maintain their IPv4 stack. Sometime in the future, exactly when no one knows, but it will probably be within the decade, IPv4 will be shut down.

Right now, to reach both the existing IPv4 sites and those that are coming on line using IPv6, you will have to have both stacks operating in your system. To add IPv6 all you have to do is

sudo apt-get install miredo

That's all. A new tap will appear in your "ip addr" listing with the name of "Teredo", and it will have an IPv6 address. That address will be this end of an IPv4-->IPv6 tunnel residing on a server in Germany or where ever. There they take four of your IPv4 packets and compose an IPv6 packet and pass it on to the IPv6 website you want to browse. Packets coming back to you through the tunnel are converted from one IPv6 packet into four IPv4 packets. So, in effect, you are still using IPv4. The ipv6 test URL mentioned in a previous post checks out your tunnel. Right now, unless you have native IPv6 hardware for your wireless, cable modem and your ISP talks IPv6, your rating will be 9/10 because of a bug in the DNS handling of Network-Manager, and IPv6 DNS resolution doesn't work right. There is a work-a-round but there is also no rush, so I am waiting till the upgrade fixes it.

@Steve, I read somewhere that with IPv6 they could give an IP address to every bullet in every magazine in every weapon on Earth. Obviously, by merely taking one from each square inch of water on the face of the planet, leaving the water with 22 left to use.

[SteveRiley]

@Steve, I read somewhere that with IPv6 they could give an IP address to every bullet in every magazine in every weapon on Earth. Obviously, by merely taking one from each square inch of water on the face of the planet, leaving the water with 22 left to use.
Huh...bullets with globally unique routable source addresses? Now there's an interesting notion...! :)

[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.

[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/ipv6-are-scandinavian-countries-really-leading-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 ? :confused:


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]

Does he refer to "pure" IPv6 ? :confused:
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 :)

[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?

[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]

....
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]

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

[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:~$

[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]

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]

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/ (http://ip-lookup.net).

[SteveRiley]

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

[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]

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.

[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? :)

[GreyGeek]

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



Jonas,
I'm still trying to understand why I don't get a 10/10, i.e., why my IPv6 DNS addresses are not honored.

What do you get when you use "route -6"?

Here's what I get while running SixXS:


:~$ route -6
Kernel IPv6 routing table
Destination Next Hop Flag Met Ref Use If
::/96 :: Un 256 0 0 sit0
2001:4978:f:580::/64 :: U 256 0 1 sixxs
fe80::/64 :: U 256 0 0 wlan0
fe80::/64 :: U 256 0 0 sixxs
::/0 2001:4978:f:580::1 UG 1024 0 0 sixxs
::/0 :: !n -1 1 140 lo
::1/128 :: Un 0 1 3 lo
::127.0.0.1/128 :: Un 0 1 0 lo
::192.168.1.101/128 :: Un 0 1 0 lo
2001:4978:f:580::2/128 :: Un 0 1 63 lo
fe80::4878:f:580:2/128 :: Un 0 1 0 lo
fe80::76de:2bff:fe36:e435/128 :: Un 0 1 0 lo
ff00::/8 :: U 256 0 0 wlan0
ff00::/8 :: U 256 0 0 sixxs
::/0 :: !n -1 1 140 lo


:: !n -1 1 131 lo


StixXS creates a gateway (UG) from default to the SixXS tunnel. Also, SixXS creates a "sit0" device, but it is down. I can bring it up with "ifconfig sit0 up", which then puts it into the route listing, but I have no idea what it is supposed to do. The tunnel works the same whether it is up or down, and with either I still get 9/10 on the ipv6test.

When I use miredo, on the other hand, there does NOT appear to be a gateway created, yet miredo still gives me a 9/10 on ipv6test, the same as sixxs. Here's the routing table with miredo loaded and functioning"


:~$ route -6
Kernel IPv6 routing table
Destination Next Hop Flag Met Ref Use If
2001::/32 :: U 256 0 0 teredo
fe80::/64 :: U 256 0 0 wlan0
fe80::/64 :: U 256 0 0 teredo
::/0 :: U 1029 0 0 teredo
::/0 :: !n -1 1 124 lo
::1/128 :: Un 0 1 3 lo
2001:0:53aa:64c:3cef:1324:e720:544/128 :: Un 0 1 0 lo
2001:4978:f:580::2/128 :: Un 0 1 53 lo
fe80::ffff:ffff:ffff/128 :: Un 0 1 0 lo
fe80::4878:f:580:2/128 :: Un 0 1 0 lo
fe80::76de:2bff:fe36:e435/128 :: Un 0 1 0 lo
ff00::/8 :: U 256 0 0 wlan0
ff00::/8 :: U 256 0 0 teredo
::/0 :: !n -1 1 124 lo


Both urls give me an extensive whois listing:


:~$ whois 2001:4978:f:580::1
#
# The following results may also be obtained via:
# http://whois.arin.net/rest/nets;q=2001:4978:f:580::1?showDetails=true&showARIN=false&ext=netref2
#




# start


NetRange: 2001:4978:: - 2001:4978:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF
CIDR: 2001:4978::/32
OriginAS:
NetName: YOURORG-NB6-AR-1
NetHandle: NET6-2001-4978-1
Parent: NET6-2001-4800-0
NetType: Direct Allocation
RegDate: 2005-10-20
Updated: 2012-03-02
Ref: http://whois.arin.net/rest/net/NET6-2001-4978-1


OrgName: YOUR.ORG, INC.
OrgId: YOURO
Address: 840 W Lake St #406
City: Roselle
StateProv: IL
PostalCode: 60172
Country: US
RegDate: 2002-12-18
Updated: 2011-09-24
Ref: http://whois.arin.net/rest/org/YOURO


OrgTechHandle: KDA6-ARIN
OrgTechName: Day, Kevin
OrgTechPhone: +1-312-884-7618
OrgTechEmail: noc@your.org
OrgTechRef: http://whois.arin.net/rest/poc/KDA6-ARIN


OrgAbuseHandle: KDA6-ARIN
OrgAbuseName: Day, Kevin
OrgAbusePhone: +1-312-884-7618
OrgAbuseEmail: noc@your.org
OrgAbuseRef: http://whois.arin.net/rest/poc/KDA6-ARIN


RTechHandle: KDA6-ARIN
RTechName: Day, Kevin
RTechPhone: +1-312-884-7618
RTechEmail: noc@your.org
RTechRef: http://whois.arin.net/rest/poc/KDA6-ARIN


RNOCHandle: KDA6-ARIN
RNOCName: Day, Kevin
RNOCPhone: +1-312-884-7618
RNOCEmail: noc@your.org
RNOCRef: http://whois.arin.net/rest/poc/KDA6-ARIN


RAbuseHandle: KDA6-ARIN
RAbuseName: Day, Kevin
RAbusePhone: +1-312-884-7618
RAbuseEmail: noc@your.org
RAbuseRef: http://whois.arin.net/rest/poc/KDA6-ARIN


# end




# start


NetRange: 2001:4978:F:: - 2001:4978:F:FFFF:FFFF:FFFF:FFFF:FFFF
CIDR: 2001:4978:F::/48
OriginAS: AS19255
NetName: YOUR-ORG-SIXXS-CHI02-TUNNELS
NetHandle: NET6-2001-4978-F-1
Parent: NET6-2001-4978-1
NetType: Reassigned
Comment: SixXS allocation for Your.org Chicago,IL PoP - This allocation is used for tunnels. - For more details, query whois.sixxs.net - More information can be found at http://www.sixxs.net/ - Abuse reports should go to abuse@sixxs.net
RegDate: 2007-10-29
Updated: 2007-10-29
Ref: http://whois.arin.net/rest/net/NET6-2001-4978-F-1


OrgName: SixXS
OrgId: SIXXS
Address: Swiss Post Box 100285
Address: Zurcherstrasse 161
City: Zurich
StateProv:
PostalCode: CH-8010
Country: CH
RegDate: 2007-10-29
Updated: 2011-11-30
Comment: Website: http://www.sixxs.net
Comment: Abuse contact: abuse@sixxs.net
Comment: Inquiries: info@sixxs.net
Ref: http://whois.arin.net/rest/org/SIXXS


ReferralServer: rwhois://whois.sixxs.net:43


OrgTechHandle: STRA1-ARIN
OrgTechName: SixXS Technical Role Account
OrgTechPhone: +41325129742
OrgTechEmail: info@sixxs.net
OrgTechRef: http://whois.arin.net/rest/poc/STRA1-ARIN


OrgAbuseHandle: STRA1-ARIN
OrgAbuseName: SixXS Technical Role Account
OrgAbusePhone: +41325129742
OrgAbuseEmail: info@sixxs.net
OrgAbuseRef: http://whois.arin.net/rest/poc/STRA1-ARIN


# end




#
# ARIN WHOIS data and services are subject to the Terms of Use
# available at: https://www.arin.net/whois_tou.html
#






Found a referral to whois.sixxs.net:43.


% This is the SixXS Whois server.
% SixXS - http://www.sixxs.net.
%
% The objects are in RPSL format.
%
% Objects not beginning with SIXXS- or ending in
% -SIXXS are cached responses from remote sources.
%
% This server outputs referto's when a object from a
% known registry (ARIN,RIPE,LACNIC,APNIC,AFRINIC,6BONE)
% is not locally available.


% Searching for prefix


inet6num: 2001:4978:f:580::/64
netname: SIXXS-USCHI02-TUN1409
descr: ayiya tunnel to 24.223.250.187


descr: Tunnel XXXXXX goes to an endpoint of JLK6-SIXXS.
country: US
remarks: Userstate: enabled
remarks: Adminstate: enabled
admin-c: JLK6-SIXXS
admin-c: KDA6-ARIN % (PoP admin-c)
tech-c: JLK6-SIXXS
tech-c: KDA6-ARIN % (PoP tech-c)
remarks: Prefixtype: Tunnel
remarks: MTU: 1280
remarks: This object is generated from the SixXS database
remarks: Abuse must be reported to abuse@sixxs.net
remarks: Information can be found at http://www.sixxs.net/
changed: info@sixxs.net 20120609
mnt-by: SIXXS-MNT
source: SIXXS


person: Kevin Day
address: P.O. 326
address: Round Lake Beach
address: IL
address: 60073
country: US
phone: +1-312-884-7618
e-mail: noc@your.org
nic-hdl: KDA6-ARIN
remarks: State: Enabled
remarks: This object is generated from the SixXS database
remarks: Abuse must be reported to abuse@sixxs.net
remarks: Information can be found at http://www.sixxs.net/
changed: info@sixxs.net 20070324
changed: info@sixxs.net 20071025
remarks: This object is partially cached into the SixXS database
remarks: and is not a 1:1 representation of the original object
source: SIXXS


person: Jerry Lynn Kreps
address: XXXXXXXXXXXXXXXXXXX
address: Lincoln, NE XXXXX
address: USA
country: US
phone: +1402XXXXXXX
e-mail: greygeek@earthlink.net
nic-hdl: JLK6-SIXXS
remarks: State: Enabled
remarks: This object is generated from the SixXS database
remarks: Abuse must be reported to abuse@sixxs.net
remarks: Information can be found at http://www.sixxs.net/
changed: info@sixxs.net 20120608
changed: info@sixxs.net 20120609
mnt-by: SIXXS-MNT
source: SIXXS


mntner: SIXXS-MNT
descr: SixXS Maintainer
admin-c: JRM1-RIPE
admin-c: PBVP1-RIPE
tech-c: PBVP1-RIPE
tech-c: JRM1-RIPE
upd-to: info@sixxs.net
mnt-nfy: info@sixxs.net
auth: CYPHERKEY XXXXXXXXXXXXXXXXXXXXXXXXXXX/
remarks: Maintains all data generated from the SixXS database.
remarks: This object is generated from the SixXS database
remarks: Abuse must be reported to abuse@sixxs.net
remarks: Information can be found at http://www.sixxs.net/
changed: info@sixxs.net 19780225
mnt-by: SIXXS-MNT
source: SIXXS


% Thank you for using the SixXS Whois Service


My end of the miredo tunnel, 2001:0:53aa:64c:3cef:1324:e720:544, just gives me a querying for my IPv4 endpoint, 24.223.250.187, of a Teredo IPv6 address, which is nothing more than my IPv4 IP address. Checking on the other end of the tunnel given in the /etc/miredo.conf file:



:~$ nslookup teredo-debian.remlab.net
Server: 127.0.0.1
Address: 127.0.0.1#53


Non-authoritative answer:
teredo-debian.remlab.net canonical name = teredo.remlab.net.
Name: teredo.remlab.net
Address: 83.170.6.76


jerry@jerry-Aspire-7739:~$ whois 83.170.6.76
% This is the RIPE Database query service.
% The objects are in RPSL format.
%
% The RIPE Database is subject to Terms and Conditions.
% See http://www.ripe.net/db/support/db-terms-conditions.pdf


% Note: this output has been filtered.
% To receive output for a database update, use the "-B" flag.


% Information related to '83.170.6.64 - 83.170.6.79'


inetnum: 83.170.6.64 - 83.170.6.79
netname: DE-IABG-TELEPORT-BSC
descr: Bernhard Schmidt
country: DE
admin-c: BSC-RIPE
tech-c: BSC-RIPE
status: ASSIGNED PA
mnt-by: IABG-MNT
mnt-lower: IABG-MNT
source: RIPE # Filtered


person: Bernhard Schmidt
address: Karl-Richter-Str. 8
address: 80939 Muenchen
address: Germany
phone: +49 89 74140493
nic-hdl: BSC-RIPE
mnt-by: IABG-MNT
source: RIPE # Filtered


% Information related to '83.170.0.0/18AS29259'


route: 83.170.0.0/18
descr: IABG mbH
origin: AS29259
mnt-lower: IABG-MNT
mnt-routes: IABG-MNT
mnt-by: IABG-MNT
source: RIPE # Filtered


% This query was served by the RIPE Database Query Service version 1.12.2 (WHOIS3)







And, another way to get routing tables:


:~$ ip route
default via 192.168.1.1 dev wlan0 proto static
169.254.0.0/16 dev wlan0 scope link metric 1000
192.168.1.0/24 dev wlan0 proto kernel scope link src 192.168.1.101 metric 2

:~$ ip -6 route
::/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
jerry@jerry-Aspire-7739:~$

[Jonas]

Here's my route -6:


jonas@kubacer:~$ route -6
Kernel IPv6 routing table
Destination Next Hop Flag Met Ref Use If
fe80::/64 :: U 256 0 0 eth1
::/0 :: !n -1 1 1 lo
::1/128 :: Un 0 1 1 lo
fe80::1af4:6aff:fe23:7b76/128 :: Un 0 1 0 lo
ff00::/8 :: U 256 0 0 eth1
::/0 :: !n -1 1 1 lo

Sorry for not being able to help more, this is all greek to me :S

I'm happy to help in anyway I can, just gimme the lines to analyse, and maybe learn a thing or two :)

b.r

jonas

[GreyGeek]

Here's my route -6:


jonas@kubacer:~$ route -6
Kernel IPv6 routing table
Destination Next Hop Flag Met Ref Use If
fe80::/64 :: U 256 0 0 eth1
::/0 :: !n -1 1 1 lo
::1/128 :: Un 0 1 1 lo
fe80::1af4:6aff:fe23:7b76/128 :: Un 0 1 0 lo
ff00::/8 :: U 256 0 0 eth1
::/0 :: !n -1 1 1 lo

Sorry for not being able to help more, this is all greek to me :S
....

So, I't pretty much greek to me as well, Jonas. I'm just tinkering around... luckily IPv6 isn't toxic or explosive.

I noticed that your setup doesn't have a UG on your IPv6 ff00::/8 eth1 connection.

And, the ip addr command shows a teredo address of 2001:0 ..... but your IPv6 routing table doesn't include it. Yet, you get out and have a 10/10 connection.



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


Mine shows I have my IPv6 sixxs device as a gateway:

::/0 2001:4978:f:580::1 UG 1024 0 0 sixxs

Apparently without a gateway and no IPv6 DNS, you communicate with IPv6 networks and have DNS resolution in them. It must all be due to your ISP.

Question:
If you do:
sudo ifconfig teredo down
or
sudo service miredo stop

do you still get 10/10?

[Jonas]

sudo ifconfig teredo down & sudo service miredo stop

Give me IPv6 result of 0/10

b.r

Jonas

[GreyGeek]

sudo ifconfig teredo down & sudo service miredo stop

Give me IPv6 result of 0/10


Ah, then your IPv6 connectivity is totally dependent on the Teredo tunnel and your DNS servers (which give you the 10th 10) are supplied by your ISP. This suggests that IF your cable modem is DOCSIS 3.0 compatible, since your ISP is IPv6 already, that:

https://wiki.ubuntu.com/IPv6#Native_IPv6_connections


Native IPv6 connections

Contact your Internet Service Provider to see if they offer IPv6 yet. Only a few Internet Service Providers (ISP) currently offer native IPv6 service, though by 2015 probably all of them will.
If your uplink involves a wifi router or broadband modem, those devices will need to support IPv6. Cable-Modems need to support DOCSIS 3.0 or 2.0+IPv6. As of 2010, very few DSL-Modems support IPv6; this will change by 2012.
If your hardware or ISP does not support native IPv6, which is the best kind, you might still be able to use IPv6 tunneled over IPv4 instead.


and you could use DHCPv6 to dispense an IPv6 address.

[Jonas]

Ah, then your IPv6 connectivity is totally dependent on the Teredo tunnel and your DNS servers (which give you the 10th 10) are supplied by your ISP. This suggests that IF your cable modem is DOCSIS 3.0 compatible, since your ISP is IPv6 already, that:

https://wiki.ubuntu.com/IPv6#Native_IPv6_connections
[FONT=courier new]

and you could use DHCPv6 to dispense an IPv6 address.

I have fiber cable - only thing I "plug in" is the router a D-Link DIR 655, which is quite a old model. Sites seem to load slower with IPv6 enabled , I've had this trouble before so I used to disable it.

b.r

Jonas

[GreyGeek]

An aside: Somehow my Chromium browser stopped preferring IPv6 over IPv4, and I't get an error message from some sites saying that I didn't have IPv6 so they couldn't display the page. I tried with FireFox it the page would display. I also "ping6"ed the websites and both ends of my tunnel and got echos back.

FireFox has a setting one can use to enable a preference for either IPv6 or IPv4. Chromium *used* to have such a setting (in the url enter "chrome://net-eternals" and on that page select "DNS". The DNS page did have a button that enabled IPv6. Now, it doesn't. Chromium now shows "ADDRESS_FAMILY_UNSPECIFIED", which causes it to consult /etc/gai.conf to determine which IP protocol to try first. It used to prever IPv6. Now it trys IPv4 first.

The man pages to gai.conf are not very extensive nor are they understandable. An example is given but no explanation. After some experimentation/trial/error I come across settings that worked to make Chromium try IPv6 first:



label ::1/128 0
label 2002::/16 1
label 2001:0::/32 2
label ::/0 3
label ::/96 4
#label ::ffff:0:0/96 5
#label fec0::/10 6
#label fc00::/7 7


#
.....
#
precedence ::1/128 50
precedence ::/0 40
precedence 2001:0::/32 35
precedence 2002::/16 30
precedence ::/96 20
#precedence ::ffff:0:0/96 10




I added the "... 2001:0::/32 2" lines to both label and precedence. My assumption is that getaddressinfo() would try them in the order of the index numbers at the end of each line. True or not, Chromium now tries the IPv6 protocol first.

[67GTA]

If you are using dd-wrt, the dev has stopped including IPV6 support. This drove me nuts for a while until I figured it out.>:(

[GreyGeek]

If you are using dd-wrt, the dev has stopped including IPV6 support. This drove me nuts for a while until I figured it out.>:(

Did the dev say why?

[67GTA]

I haven't found any official statement. I ran across a post on their forums where every one was asking where IPV6 had gone. All of the older builds had support, but was dropped somewhere down the line. You are supposed to be able to set it up, but my Linksys WRT160n doesn't have the ram. I already have to use the mini build to run dd-wrt on it. http://www.dd-wrt.com/wiki/index.php/IPv6

[SteveRiley]

What about OpenWRT? Might be a suitable alternative.

[GreyGeek]

What about OpenWRT? Might be a suitable alternative.

Apparently the same way I get it with Kubuntu: http://wiki.openwrt.org/doc/howto/ipv6


Obtain IPv6 supportFollow ipv6.essentials (http://wiki.openwrt.org/doc/howto/ipv6.essentials) to obtain full IPv6 support. Then come back and read about the configuration here:
There are two big, different steps:


Set up a working IPv6 connection on the OpenWrt router, either by tunneling (SixXs, TSP, 6to4), or natively

Propagate the IPv6 subnet to the LAN with RADVD (http://wiki.openwrt.org/doc/howto/ipv6#radvd) or DHCPv6 (http://wiki.openwrt.org/doc/howto/ipv6#dhcpv6).


Native IPv6 accessFor this, you need to obtain an IPv6 address from your ISP. Technically this could be a /128 prefix (exactly one IPv6 address), but according to rfc6177 (http://tools.ietf.org/html/rfc6177) this should be a /64 prefix. You may also get bigger range, like /56 or /48. Within this range you may use all the IPv6 addresses to your liking without any NAT-induced headaches.



Here is someone's attempt: https://forum.openwrt.org/viewtopic.php?id=27541


I was looking for information about configuring my router (originally Netgear WNDR3700) for supporting a static 6in4 tunnel from SixXS for getting the IPv6 connectivity. The router is currently running OpenWrt Backfire 10.03.1-rc4, published in November 2010, the rc4 version of the forthcoming Backfire Interim Release 1.

Current OpenWrt Backfire 10.03.1-rc4 has an installable package for supporting the 6in4 tunnels, which should make the process relatively easy. However, finding the exactly correct configuration is not that easy. I found useful information in internet, but it was scattered around and to some extent also outdated or incomplete. I write this article to summarize my findings and to list my configuration as an example for others trying to do the same.

Additionally, the rc4 version does not enable configuring some of the required steps through the Luci GUI, so some tasks have to be done by editing configuration files manually.

Background assumptions: you have a "Static" 6in4 tunnel with a fixed tunnel endpoint from SixXS. You also have a subnet, which is routed through that tunnel. You also have installed the OpenWrt to the router.

Main steps in the process:
Configuring the tunnel
Configuring iptables to make sure that the tunnel stays up
Configuring IPv6 address autoconfiguration inside local LAN by using RADVD
Configuring ipv6 firewall - ip6tables
...

... and it goes on. With that advice and installing miredo I can get a dual stack. ;D

[GreyGeek]

If you are using dd-wrt, the dev has stopped including IPV6 support. This drove me nuts for a while until I figured it out.>:(

I did some searching and found this:http://www.dd-wrt.com/wiki/index.php/What_is_DD-WRT%3F#Notes


(5)Apparently, IPv6-related features DO NOT work by default in DD-WRT v24. See IPv6 on v24 (http://www.dd-wrt.com/wiki/index.php/IPv6_on_v24).


That link led to this one:
http://www.dd-wrt.com/wiki/index.php/IPv6_on_v24
This link states:


IPv6 On V24
UPDATE: If you are just trying to get IPv6 working (6to4) on v24 sp1, please view this page:
http://www.dd-wrt.com/wiki/index.php/IPv6#6to4_Setup

...
The currently recommended K2.6 big images ship with basic IPv6 support. That means that you're able to activate the IPv6 IP stack for routing and static tunnels. What's missing are the ip6tables kernel modules and the userland tools needed. It's possible to add the missing parts but that requires some basic Linux system knowledge. It's assumed here that you have /jffs mounted and a few megs of space available and you know how to use ipkg.Please note: What you've to download still depends a lot on the hardware you're running on!

Kernel modulesThe currently recommended build r14929 ships with a patched Linux kernel with a version number set to 2.6.24.111. Since the kernel modules from OpenWRT have the magic version information set you can't use the kmod packages from OpenWRT. Because of the version mismatch insmod won't load those modules.
This leaves you with two options:


Trust lazytom and download his set of modules from the forum http://www.dd-wrt.com/phpBB2/download.php?id=16285
If you don't trust random people on the internet (which is a good thing) you'll find an introduction on how to build these modules yourself in this blog post http://blog.dest-unreach.be/2010/12/01/compiling-custom-dd-wrt-kernel-modules

.....


Which leads to:


6to4 (http://en.wikipedia.org/wiki/6to4) is a mechanism which assigns a large block of IPv6 addresses to every IPv4 address on the Internet. You can use 6to4 with DD-WRT to give every computer on your network a globally-accessible IPv6 address without the need to configure explicit tunneling. This should be the simplest method to add IPv6 to your entire network.
Important Reminders :


In order to use ipv6, you need the Standard or VoIP version of DD-WRT, as these are currently the only ones which support both IPv6 and radvd.
The detailled configuration steps are targeted toward users with a basic DHCP connection for the WAN part. So, if using PPPoE will require replacing vlan1 with ppp0 in each instance. Other connection types will vary.
When using DD-WRT standard on a router with 4MB Flash, there is no space available for jffs.
This guide only relies on nvram variables, so that jffs is not needed.
The configuration shown below has been verified to work properly on v24 sp1 stable (standard)

...


and we are back to tunnels again.

[SteveRiley]

We're living in an IPv4-IPv6 transition world right now. It will be almost impossible to avoid tunnels until IPv6 is running natively everywhere.

[67GTA]

I've looked around, and I think only tomato has full ipv6 support, but you still have to fiddle with it. It wasn't worth the trouble to me. I just added this to the list of "Honey, this is why I need a new router".