DC it seems.. specifically #13 in the partial trace: 11 cr2.n54ny.ip.att.net (12.122.130.102) 87.720 ms 85.642 ms 87.083 ms 12 cr2.wswdc.ip.att.net (12.122.3.38) 85.815 ms 89.022 ms 89.031 ms 13 12.122.134.153 (12.122.134.153) 88.434 ms 85.743 ms 85.440 ms and #13 in this one: 9 xe-0-1-0.er1.dca2.us.above.net (64.125.27.25) 103.117 ms 103.171 ms 103.140 ms 10 xe-1-1-0.er1.iad10.above.net (64.125.26.238) 81.522 ms 81.509 ms 81.570 ms 11 192.205.36.125 (192.205.36.125) 82.556 ms 82.594 ms 82.570 ms 12 cr1.wswdc.ip.att.net (12.122.82.222) 85.047 ms cr2.wswdc.ip.att.net (12.122.84.46) 84.959 ms 84.784 ms 13 12.122.134.1 (12.122.134.1) 84.224 ms 84.719 ms 12.122.135.1 (12.122.135.1) 84.745 ms On Thu, 2009-11-05 at 09:21 -0800, Ali S wrote:

Washington state or DC ?

sent via droid...

...

On Nov 5, 2009 8:25 AM, "Leland Vandervort" <leland@taranta.discpro.org> wrote:

Hi all,

Anyone have any information on the 20 minute (or so) flap that appears to have occured in Washington between AT&T and the rest of the world?

We saw a complete outage of connectivity to several APIs at a few parnters who happen to all be behind AT&T. Although in BGP the routes were still being advertised across this peering from AT&T to the world, traffic just died at the washington peering with AT&T irrespective of which path we took. (3 different transits here).

Just curious.

Thanks,

Leland