Trouble reaching whois.ripe.net and www.ietf.org

James Jun james at towardex.com
Tue Oct 17 18:33:18 CEST 2006


>From occaid prespective, there is alternate path available to reach 2603 at
Stockholm, but it seems L(3) is doing just fine from here, at least to the
next-hop shown in your BGP show commands.

It may be possible it's not working for IETF's crit-infrastructure /48.  If
they can run a traceroute for us to 2001:948:0:F042::1 that might help
better.


traceroute6 to 2001:948:0:F042::1 (2001:948:0:f042::1), 64 hops max, 12 byte
packets
 1  0.fe1-3.cr1.atl1.us.occaid.net  1.024 ms  0.853 ms  0.689 ms
 2  ge-0-1-0.cr1.iad1.us.occaid.net  14.565 ms  14.537 ms  14.431 ms
 3  21.ge0-0.cr1.ewr1.us.occaid.net  20.220 ms  20.237 ms  20.935 ms
 4  v3323-mpd.cr1.lhr1.uk.occaid.net  90.316 ms  89.977 ms  89.899 ms
 5  2001:7f8:4::d1c:1  91.032 ms  90.645 ms  90.505 ms
 6  2001:1900:5:2::6  125.838 ms  126.067 ms  126.017 ms
 7  so-7-2-0.rt1.vie.at.geant2.net  126.020 ms  125.966 ms  125.944 ms
 8  so-6-3-0.rt1.fra.de.geant2.net  126.554 ms  126.044 ms  126.144 ms
 9  so-1-3-0.rt1.lux.lu.geant2.net  139.965 ms  139.980 ms  139.912 ms
10  nordunet-gw.rt1.cop.dk.geant2.net  140.304 ms  140.138 ms  140.378 ms
11  dk-gw.nordu.net  140.489 ms dk-gw.nordu.net  140.535 ms dk-gw.nordu.net
140.778 ms
12  no-gw2.nordu.net  148.424 ms  148.354 ms  148.296 ms
13  no-gw.nordu.net  148.577 ms  149.466 ms  148.467 ms

James

> -----Original Message-----
> For www.ietf.org, the prefix 2001:503:C779::/48 is used and the story
> is similar. We get an in-optimal path, and it again seems to stop in
> level3.
> 
> *> 2001:503:C779::/48
>                      2001:948:0:F042::1
>                                                             0 2603 20965
> 3356 30071 7786 i
> 
> Use of these longer prefixes for critical infrastructure seems like a
> bad idea to me. Due to many people filtering these, the paths are at
> best in-optimal, and in this case not working at all.
> 
> I assume you guys see similar things?
> 
> Stig




More information about the ipv6-ops mailing list