nanog mailing list archives
Re: Geographic map of IPv6 availability
From: Kevin Day <toasty () dragondata com>
Date: Fri, 5 Oct 2007 23:53:13 -0500
On Oct 5, 2007, at 10:38 PM, Kevin Loch wrote:
Nathan Ward wrote:On 6/10/2007, at 3:18 AM, Stephen Wilcox wrote:That's because the 'v6 network' is broken enough that putting AAAA records on sites that need to be well reachable is a bad idea. For example, due mainly to Vista's 6to4 tunnelling stuff (based on researching a random sample of users), I'd lose about 4% of visitors to my web-sites if I were to turn on AAAA records.<stuff> Given the above, I think there is no myth.. !Has anyone who was using AAAA records for a site turned them off due toreachability problems?
Yes. We tried it on one of our client's rather high profile sites and had to turn it off because of exactly that problem.
We found a few friendly tech savvy users who were experiencing the problem and followed up with them the best we could. The reasons for the problems were:
Some had inadvertently enabled 6to4 (one admitted remembering playing with it after reading about it on slashdot, then forgot about it). Some had installed one vendor's firewall that was trying to be proactive and firewall v6 things as well. We never determined if this was default behavior or not, but if you checked the "Firewall v6 traffic" box, it enabled the whole v6 stack just so that it could firewall it. Some we were never able to figure out why it broke connectivity for them. Theories about transparent proxies doing the wrong thing, broken resolvers or other issues floated up, but we could never pin them down.
It definitely wasn't in the order of entire percentage points of users being unable to access the site, but it was a non-zero number and enough to make the site owner want the AAAA records pulled.
I talked about this briefly at http://www.ipv6experiment.com and it's one of the things we plan on trying to measure when we finally get it up and running.
-- Kevin
Current thread:
- Geographic map of IPv6 availability michael.dillon (Oct 03)
- Re: Geographic map of IPv6 availability Rik van Riel (Oct 05)
- Re: Geographic map of IPv6 availability Marshall Eubanks (Oct 05)
- RE: Geographic map of IPv6 availability michael.dillon (Oct 05)
- Re: Geographic map of IPv6 availability Marshall Eubanks (Oct 05)
- Re: Geographic map of IPv6 availability Stephen Wilcox (Oct 05)
- Re: Geographic map of IPv6 availability Nathan Ward (Oct 05)
- Re: Geographic map of IPv6 availability Kevin Loch (Oct 05)
- Re: Geographic map of IPv6 availability Kevin Day (Oct 05)
- Re: Geographic map of IPv6 availability Stephen Wilcox (Oct 06)
- Re: Geographic map of IPv6 availability Peter Dambier (Oct 06)
- Re: Geographic map of IPv6 availability Joe Abley (Oct 06)
- Re: Geographic map of IPv6 availability Stephen Wilcox (Oct 06)
- Re: Geographic map of IPv6 availability Joe Abley (Oct 06)
- Re: Geographic map of IPv6 availability Nathan Ward (Oct 05)
- Re: Geographic map of IPv6 availability Rik van Riel (Oct 05)
- Re: Geographic map of IPv6 availability Mark Prior (Oct 07)
- RE: Geographic map of IPv6 availability Tony Hain (Oct 11)
- Re: Geographic map of IPv6 availability Kevin Loch (Oct 11)
- Re: Geographic map of IPv6 availability Nathan Ward (Oct 11)
- <Possible follow-ups>
- Re: Geographic map of IPv6 availability Paul Vixie (Oct 13)
