http://www.google.com/intl/en/ipv6/
http://www.networkworld.com/news/2009/032509-google-ipv6-easy.html
Any one making use of Google IPV6?
Robert D. Scott Robert@ufl.edu
Senior Network Engineer 352-273-0113 Phone
CNS - Network Services 352-392-2061 CNS Phone Tree
University of Florida 352-392-9440 FAX
Florida Lambda Rail 352-294-3571 FLR NOC
Gainesville, FL 32611 321-663-0421 Cell
Yes I do.
I can use it but sometimes got trouble with teredo.
Retry half an hour later works 
ipv6.google.com looks better to me than the IPv4 version does.
More comfort. It is worth the trouble with teredo.
Peter
Robert D. Scott wrote:
yup... and it is nice, adwords don't work pretty well (or at least on the GeoIP thingie), and i get less publicity to look at
Robert D. Scott wrote:
http://www.google.com/intl/en/ipv6/
http://www.networkworld.com/news/2009/032509-google-ipv6-easy.html
Any one making use of Google IPV6?
It's been my Firefox home page ever since it was available.
Steve
yes. We participate in the Google IPv6 trial program so our recursors
get AAAA records for www.google.com and so far it's been great, no
issues whatsoever.
daniel@jun1> traceroute www.google.com
traceroute6 to www.l.google.com (2001:4860:a003::68) from
2001:7f8:1::a501:2859:2, 64 hops max, 12 byte packets
1 pr61.ams04.net.google.com (2001:7f8:1::a501:5169:1) 2.388 ms 1.798
ms 1.712 ms
2 2001:4860::23 (2001:4860::23) 8.664 ms 8.480 ms 8.364 ms
3 2001:4860:a003::68 (2001:4860:a003::68) 8.624 ms 8.639 ms 8.719
ms
Regards,
Daniel.
Um, are you sure you are using Google over IPv6?
This is *not the same thing* as ipv6.google.com.
Google over IPv6 is about accessing www.google.com via IPv6. For you to
be doing this, you must have IPv6 connectivity and your IPv6 network
must meet Google's fairly stringent requirements.
Regards, K.
Same.
We've been participating since January and haven't had any problems:
# traceroute6 www.google.com
traceroute to www.google.com (2001:4860:c003::68), 30 hops max, 40 byte
packets
1 vl2-gw.cbr1.as24557.net.au (2405:5000:1:2::1) 0.492 ms 0.484 ms 0.501
ms
2 gi0-1-4.bdr1.syd1.as24557.net.au (2405:5000:1:4::21) 5.009 ms 5.048 ms
5.212 ms
3 AS15169.ipv6.sydney.pipenetworks.com (2001:7fa:b::14) 4.552 ms 4.538
ms 4.522 ms
4 2001:4860::29 (2001:4860::29) 157.930 ms 157.914 ms 149.638 ms
5 2001:4860:c003::68 (2001:4860:c003::68) 157.709 ms 156.651 ms 149.585
ms
-Shaun
It's working for me, too, though I noticed that tcptraceroute (at least
the version I have) doesn't do well with ipv6.google.com.
--Steve Bellovin, http://www.cs.columbia.edu/~smb
seems to work fine from over here:
# tcptraceroute6 www.google.com 80
traceroute to www.google.com (2001:4860:a003::68) from
2001:7b8:3:30::<removed>, port 80, from port 62699, 30 hops max, 60 byte
packets
1 2001:7b8:3:30::2 (2001:7b8:3:30::2) 0.505 ms 0.246 ms 0.228 ms
2 pr61.ams04.net.google.com (2001:7f8:1::a501:5169:1) 1.664 ms 1.619
ms 1.641 ms
3 2001:4860::23 (2001:4860::23) 220.972 ms 174.560 ms 120.445 ms
4 2001:4860:a003::68 (2001:4860:a003::68) 9.101 ms [open] 9.196 ms
9.055 ms
# tcptraceroute6 -V
traceroute6: TCP & UDP IPv6 traceroute tool 0.9.3 ($Rev: 483 $)
--Daniel.
Traceroute6 works; I'm talking about tcptraceroute, which is useful for
seeing what happens to connections in the presence of ACLs, firewalls,
and the like. I don't seem to have a tcptraceroute6.
--Steve Bellovin, http://www.cs.columbia.edu/~smb
In a message written on Fri, Mar 27, 2009 at 08:18:50AM -0400, Robert D. Scott wrote:
Any one making use of Google IPV6?
We are in the trial:
% traceroute6 -n www.google.com
traceroute6 to www.l.google.com (2001:4860:b002::68) from
2001:4f8:3:bb::5, 64 hops max, 12 byte packets
1 2001:4f8:3:bb:203:47ff:fefd:ddab 0.350 ms 0.263 ms 0.233 ms
2 2001:4f8:3:c::1 0.734 ms 0.481 ms 0.613 ms
3 2001:4f8:0:4::3 4.602 ms 5.859 ms 5.234 ms
4 2001:4f8:0:1::45:1 78.206 ms 2.829 ms 1.858 ms
5 2001:504:d::1f 13.601 ms 1.607 ms 1.738 ms
6 2001:4860::30 80.442 ms 70.358 ms 68.369 ms
7 2001:4860:b002::68 70.092 ms 68.064 ms 68.021 ms
Completely seamless from here.
Daniel Verlouw wrote:
yes. We participate in the Google IPv6 trial program so our recursors
get AAAA records for www.google.com and so far it's been great, no
issues whatsoever.
Same experiences - it just works.
daniel@jun1> traceroute www.google.com traceroute6 to www.l.google.com (2001:4860:a003::68) from
2001:7f8:1::a501:2859:2, 64 hops max, 12 byte packets
1 pr61.ams04.net.google.com (2001:7f8:1::a501:5169:1) 2.388 ms 1.798
ms 1.712 ms
2 2001:4860::23 (2001:4860::23) 8.664 ms 8.480 ms 8.364 ms
3 2001:4860:a003::68 (2001:4860:a003::68) 8.624 ms 8.639 ms 8.719
ms
Yes, but only www records have AAAA record, the domain (google.com without www prefix) is still IPv4 only.
Their press would indicate that more than www is IPV6.
When I posted my original note, I was not really looking for end user
feedback, but rather is anyone peering V6 with them on either a public
fabric or private peer. Any idea if they have native V6 transit, or are
tunneling, and to where.
Robert D. Scott Robert@ufl.edu
Senior Network Engineer 352-273-0113 Phone
CNS - Network Services 352-392-2061 CNS Phone Tree
University of Florida 352-392-9440 FAX
Florida Lambda Rail 352-294-3571 FLR NOC
Gainesville, FL 32611 321-663-0421 Cell
Robert D. Scott wrote:
When I posted my original note, I was not really looking for end user
feedback, but rather is anyone peering V6 with them on either a public
fabric or private peer. Any idea if they have native V6 transit, or are
tunneling, and to where.
No tuneling I think. We have with them several peerings, IPv6 native together with IPv4.
When I posted my original note, I was not really looking for end user
feedback, but rather is anyone peering V6 with them on either a public
fabric or private peer. Any idea if they have native V6 transit, or are
tunneling, and to where.
They are peering over some IXPs and private peerings with native IPv6,
and I believe Google like to check IPv6 connectivity before putting
your DNS resolver addresses in a whitelist so AAAA records are
returned.
Regards,
Rob
In a message written on Fri, Mar 27, 2009 at 11:03:05AM -0400, Robert D. Scott wrote:
When I posted my original note, I was not really looking for end user
feedback, but rather is anyone peering V6 with them on either a public
fabric or private peer. Any idea if they have native V6 transit, or are
tunneling, and to where.
AFAIK you have to have native peering with them to be part of the
pilot. At least, you did when we signed up. They may have relaxed
that since.
Heard that they are somewhat picky about who they AAAA-enable. Our campus
has had native IPv6 everywhere and upwards all the way to Geant for many
years. We are thinking of applying in the hopes that it will boost IPv6
usage. Did you have any trouble getting them to IPv6-enable you? Anyone
from Google in the list with any informative comment?
Regards,
Athanasios
According to a Google IPv6 talk I attended yesterday, they don't intend to relax that rule. Tunneling ipv6 connectivity over ipv4 is trash quality engineering and to be honest, its not a credible substitute for adequate ipv6 infrastructure.
Nick
Steven M. Bellovin wrote:
Robert D. Scott wrote:
http://www.google.com/intl/en/ipv6/
http://www.networkworld.com/news/2009/032509-google-ipv6-easy.html
It's relatively easy to make _your own_ apps (i.e. ones you have the source for) support IPv6.
Most companies, though, are completely reliant on their vendors, which means buying a new version, testing, deployment, etc. -- assuming the vendor is still in business, hasn't discontinued the product, has even bothered to try implementing IPv6 yet (most haven't), etc. That may also involve an upgrade of the OS that the app runs on, purchasing new hardware to handle the bloat in newer OSes, etc. You may also need to upgrade your LAN hardware to models that support IPv6 forwarding in hardware, more RAM for routers to run IPv6 code (if it's even available), new VPN boxes, etc.
Now, if you keep up with your upgrades every year, and stop using products when the vendors stop supporting them or go out of business, most of this should already be built into your budgets -- but not many execs see value in that. "If it ain't broke so badly that it cuts into profits, you don't need any budget for it."
S