[123461] in North American Network Operators' Group
IPv6 client adoption slowly going 3% (?)
daemon@ATHENA.MIT.EDU (=?iso-8859-1?Q?Ask_Bj=F8rn_Hansen?)
Tue Mar 9 14:46:07 2010
From: =?iso-8859-1?Q?Ask_Bj=F8rn_Hansen?= <ask@develooper.com>
Date: Tue, 9 Mar 2010 00:38:58 -0800
To: NANOG list <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
Hi everyone,
I read the IPv4 address space thread and thought this might be of =
interest to the group:
Some months ago I setup a small test to test if users browsing =
http://www.pool.ntp.org/ supported IPv6:
http://www.v6test.develooper.com/statistics
http://www.v6test.develooper.com/
It's slowly been creeping up to 2.8% this month (from ~2.2% in =
November). At this rate we'll have critical mass (say 25%?) already in =
2022!
Kidding aside, although most of the test data is from NTP Pool website =
currently I've been testing a few different sites and the numbers are =
all pretty close.
Anyway - if you want to test your user base, then you can make an =
account at http://www.v6test.develooper.com/account and get a bit of =
javascript to put on your website/weblog/...
The code for the system is at=20
http://git.develooper.com/?p=3Dv6test.git;a=3Dsummary and =
http://github.com/abh/v6test
Specifically the javascript doing the test at:
http://github.com/abh/v6test/blob/master/public/devel/v6test.js
I've also been testing how many fails to handle A+AAAA records and it =
seems to be about 2-3%[1] -- which sounds like more than plenty for =
Google, Amazon etc not to enable IPv6 on their regular hostnames where =
it matters.
- ask
[1] The number on the "global" statistics page is artificially low =
because many of the tests have been via a host with both A and AAAA =
records and I forgot to take that into account when calculating the =
numbers. Oops. I'll get it fixed.
--=20
http://develooper.com/ - http://askask.com/
