IPv6 Information Wiki

ARIN has set up a wiki at http://www.getipv6.info to publish information
that will help ISPs, large and small in implementing IPv6 and migrating
to an IPv6 Internet.

I have collected info from some recent emails to the NANOG list, and
other sources, and published it on the wiki. Since this is a wiki,
anyone with something useful can contribute. In particular, it would be
good for more eyes to review the wiki for technical correctness and for
more people to participate in discussing IPv6 addressing plans at
http://www.getipv6.info/index.php/IPv6_Addressing_Plans

Thanks,

--Michael Dillon

<michael.dillon@bt.com> writes:

ARIN has set up a wiki at http://www.getipv6.info to publish information
that will help ISPs, large and small in implementing IPv6 and migrating
to an IPv6 Internet.

The unintentionally funny part of this is that the wiki hangs (the
redirect to ARIN IPv6 Wiki - ARIN's Vault from
http://www.getipv6.info works, but the subsequent page load does not)
if you try to connect to it using IPv6. Is this a subtle way of
saying that I am not the intended audience?

                                        ---Rob

[added bcc to ops@arin.net so that they can have a look at it too from
their end etc etc]

Robert E. Seastrom wrote:

<michael.dillon@bt.com> writes:

ARIN has set up a wiki at http://www.getipv6.info to publish information
that will help ISPs, large and small in implementing IPv6 and migrating
to an IPv6 Internet.

The unintentionally funny part of this is that the wiki hangs (the
redirect to ARIN IPv6 Wiki - ARIN's Vault from
http://www.getipv6.info works, but the subsequent page load does not)
if you try to connect to it using IPv6. Is this a subtle way of
saying that I am not the intended audience?

Welcome to the wonderful world of "MTU issues" :wink:

$ tracepath6 www.getipv6.info
1?: [LOCALHOST] pmtu 1500
1: ge-1-3-0.breda.ipv6.concepts-ict.net asymm 2 1.320ms
2: 2001:838:0:10::1 asymm 3 3.404ms
3: ge6-2-0.br0.ams3.nl.gbxs.net asymm 5 4.666ms
4: so-0-0-0.bb1.bru2.be.gbxs.net asymm 5 8.504ms
5: ge-0-3-0.bb1.bru1.be.gbxs.net asymm 6 8.662ms
6: ge-6-0-0-32.bb1.bru2.be.gbxs.net asymm 8 15.684ms
7: 2001:7f8:4::cb9:1 asymm 9 15.809ms
8: so-1-0-0.dus11.ip6.tiscali.net 17.379ms
9: so-0-0-0.ham10.ip6.tiscali.net 23.015ms
10: so-0-0-0.ham10.ip6.tiscali.net asymm 9 22.305ms pmtu 1480
10: sl-bb1v6-bru-t-7.sprintv6.net asymm 9 38.707ms
11: sl-bb1v6-rly-t-1001.sprintv6.net asymm 10 148.863ms
12: 2001:440:1239:7000::2 asymm 11 155.694ms
13: no reply
14: no reply
15: no reply

Something is nicely filtering out ICMP there and clearly some hop is not
1480 :wink: The fun really starts when you have a 1280 pMTU though.

Unfortunately ARIN doesn't have any traceroute6/tracepath6 tool
available online so that one can check the path back, which could show
where it goes wrong, as asymetric routes tend to be a big issue also :wink:

I won't even show the other traceroute which has a 400ms+ path and
bounces from between nl -> uk -> us -> jp -> kr -> us -> arin and then
goes dead. IMHO ARIN should also definitely look at getting a good
transit provider, which would also help quite a bit with these kind of
problems.

Greets,
Jeroen

It might be worth syncing up with the people who are working on <http://ipv6.cluepon.net/>, in the interests of concentrating effort.

Joe

ARIN has set up a wiki at http://www.getipv6.info to publish
information that will help ISPs, large and small in implementing
IPv6 and migrating to an IPv6 Internet.

It might be worth syncing up with the people who are working on
<http://ipv6.cluepon.net/&gt;, in the interests of concentrating effort.

or the folk working on <http://www.civil-tongue.net/clusterf/&gt;\. there
seem to be more folk working on v6 wikis than vendors working on fully
functional v6/dual-stack implementations.

randy

at least we'll have good documentation of what people aren't implementing. :slight_smile:

or the folk working on <http://www.civil-tongue.net/clusterf/&gt;\. there
seem to be more folk working on v6 wikis than vendors working on fully
functional v6/dual-stack implementations.

at least we'll have good documentation of what people aren't implementing. :slight_smile:

which one hopes will allow us to focus on getting them to do so.
pretending everything is just lovely a la jordi sure has not done it.

randy