Operational: Wiltel Peering with MCI problems around D.C

Richard A Steenbergen ras at e-gerbil.net
Thu Aug 18 20:13:27 UTC 2005


On Thu, Aug 18, 2005 at 03:41:56PM -0400, Rich Emmings wrote:
> 
> Anyone else (Wiltel customers especially) running into an operational issue 
> around D.C. with partial connectivity
> 
> It would seem MCI and Wiltel around D.C. have a 'informal' peering 
> relationship and it's been errored right now for about 39 hours with a 
> half-duplex route announcement.  This has been effecting us with some loss 

Informal peering relationship with a half duplex route announcement? Is 
this some really obscure way of saying someone is leaking routes? Trust 
me, there is no such thing as an "informal" peering relationship with MCI.

> of connectivity that's not there when we test same sites from other ISP 
> clouds.  Since it's informal, the help desk system at one or both ands may 
> be having problem entering a ticket w/o an account number for the circuit.

I don't think so. Whether it is a peer or a transit relationship, nothing 
that big between those parties is informal, undocumented, or unticketable. 
Besides, in this case, WCG buys transit from MCI:

GigabitEthernet5-0.GW4.IAD8.ALTER.NET (157.130.30.245)

Name:    wcgGigE-gw2.customer.alter.net
Address:  157.130.30.246

I will take your word tht something is broken though, as the traceroute 
from Washington DC to www.mci.com via http://lookingglass.wcg.net is:

Tracing the route to global.mci.com (164.109.35.20)

1 so-0-0-0.edge2.Washington1.Level3.net (4.68.127.25) [AS 3356] 0 msec 0 msec 0 msec
2 * * * 
3 * * *
...

> The usual channels are not producting results, and we're starting to get 
> engineers on the lower end of the evoluationary food chain and finger 
> pointing between wgc & mci that's not helping.  Tried a pch, haven't heard 
> yet.

Are you actually a customer of either one? If so, you ring them up and 
stay on the phone until they address the problem. If not, take a stab at 
it that WCG is the most interested party in getting it fixed, and try 
nagging them some more. :)

-- 
Richard A Steenbergen <ras at e-gerbil.net>       http://www.e-gerbil.net/ras
GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)



More information about the NANOG mailing list