Lars Eggert wrote:
Tom Marshall wrote:
I have found that IPv6 seems to have a much longer round
trip delay than
IPv4 for just about every site that I have tried. For
example, pinging
www.rfc-editor.org from here gives me this:
IPv6: rtt min/avg/max/mdev = 589.174/678.001/1109.953/117.776 ms
IPv4: rtt min/avg/max/mdev = 8.346/48.524/130.290/37.593 ms
Also, note that I cannot get an HTTP response from
www.rfc-editor.org using
IPv6. The connection is established fine, but the server does not
respond
to requests.
I can't comment on the routing issues (IPv6 connectivity esp. to the
6bone seems to have been flaky), but the RFC editor IPv6 server has
been
up since last week.
You should see "you are using IPv6 from
2001:490:f002:128:240:96ff:fe40:5bc9" on the bottom of the front page
if
all went well.
This page was last updated on 12Aug02.
You are using IPv6 from 3ffe:8114:2000:240:200:39ff:fe77:1f3f
Thus IE6/Win2k works and I expect most others supporting IPv6 to do so
also ;)
As for latency:
IPv6: 10 packets transmitted, 10 received, 0% packet loss, time 9097ms
rtt min/avg/max/mdev = 174.092/175.407/178.517/1.313 ms
IPv4: 10 packets transmitted, 10 received, 0% packet loss, time 9086ms
rtt min/avg/max/mdev = 187.186/190.781/199.164/3.799 ms
Looks the same to me, IPv6 is even a fraction better.. hmm ;)
<lobby>
Shall I add that host, or one near it, to:
http://www.sixxs.net/misc/latency/ ?
Routing glitches can be seen quite well in that system.
Btw, is there a ISI Looking Glass or similar system available
so that it can be included into GRH (http://www.sixxs.net/tools/grh/) ?
Check the LG on the last url shows that ASN4554 does quite some
transit even between europe -> us and back to europe again...
Ofcourse everybody else with an AS who like to share
their view onto the world can take a look.
</lobby>
Greets,
Jeroen