www.arin.net

[Steve Powell]

Oops, I better check the archive if there are any other reported
GBLX issues.  2001:500:4::/48 should be fixed.  The tunnel was
busted between our CDG router and the legacy overlay IPv6 router 
in Chicago.  I have observed a couple corner case problems running
6PE on the legacy overlay routers.  Generally a reset of a tunnel
fixes the problem.  

spowell at ar2.cdg2# run traceroute 2001:500:4::1
traceroute6 to 2001:500:4::1 (2001:500:4::1) from 2001:450:2001:1000:0:670:1708:128, 30 hops max, 12 byte packets
 1  2001:450:1:2000::c (2001:450:1:2000::c)  103.510 ms  103.573 ms  103.164 ms
 2  2001:450:1:2000::c (2001:450:1:2000::c)  103.278 ms  103.293 ms  104.041 ms
 3  2001:450:1:1::29 (2001:450:1:1::29)  103.881 ms  103.396 ms  103.404 ms
 4  2001:450:1:2000::d (2001:450:1:2000::d)  104.364 ms  121.718 ms  143.749 ms
 5  ge-0-1-0.cr1.iad1.us.occaid.net (2001:4830:ff:f151::2)  131.699 ms  131.526 ms  131.223 ms
 6  arin-gw1.customer.occaid.net (2001:4830:e6:a::2)  132.508 ms  132.999 ms  132.291 ms
 7  arin-gw1.customer.occaid.net (2001:4830:e6:a::2)  3129.606 ms !A  3132.312 ms !A  3145.123 ms !A


Ashburn Equinix is a known issue.  We have a 7200 hanging off of the 
IPv6 exchange which is legacy.  We have been trying to pull the 7200 and rehome
directly to a production router.  But there were a couple customers
we had to move first.  Looks like the customers were moved.  Let
me see if I can get that 7200 pulled out and rehomed to a production
router.  That should resolve the MTU issues as I'm presuming the
Equinix switch can handle a larger MTU. 

stevep