[26249] in North American Network Operators' Group
Re: BBN issues
daemon@ATHENA.MIT.EDU (Danny McPherson)
Mon Dec 13 19:13:55 1999
Message-Id: <199912140014.RAA17711@ice.ip.qwest.net>
To: nanog@merit.edu
From: Danny McPherson <danny@qwest.net>
Reply-To: danny@ice.ip.qwest.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Mon, 13 Dec 1999 17:14:50 -0700
Errors-To: owner-nanog-outgoing@merit.edu
> You don't see the NY -> Cali -> NY path? That is a bad thing (tm).
In the first traceroute it did initially seem interesting, although my
presumption was that Sprint likely wasn't leaking routes (or providing
transit?) "everywhere", and probably only via that specific path, as
demonstrated. Another observation to that regard is that the [leaked?] prefix
was presumably a more specific, else [theoretically], Genuity would have
selected the shorter AS path .. although BBN/GTEI may perhaps be providing
that transit via there AS, which would even things out from a path vector
perspective.
(The reverse path seems somewhat reasonable considering BBN and Genuity are
still separate ASs with distinct network and interconnection topologies.
I do still see potential brokenness wrt route leakage and the like, so either Sprint is still leaking routes or GTEI is attempting to offload (via a psuedo transit connection) some of that Genuity hosting traffic elsewhere...
Of course, I'd really suggest now that we avoid dicussing this specifically on the list and leave the problem resolution to the folks that can actually fix it/them, if they actually exist.
-danny
