unavailable virtual gateways

Robert Woody Woodburn <woody@sparta.com> Tue, 26 May 1992 17:51 UTC

Received: from nri.nri.reston.va.us by ietf.NRI.Reston.VA.US id aa02255; 26 May 92 13:51 EDT
Received: from nri.reston.va.us by NRI.Reston.VA.US id aa25617; 26 May 92 13:58 EDT
Received: from PIZZA.BBN.COM by NRI.Reston.VA.US id aa25613; 26 May 92 13:58 EDT
Received: from pizza by PIZZA.BBN.COM id aa01503; 26 May 92 13:46 EDT
Received: from BBN.COM by PIZZA.BBN.COM id aa01494; 26 May 92 13:44 EDT
Received: from SPARTA.COM by BBN.COM id aa01907; 26 May 92 12:39 EDT
Received: by sparta.com (5.65/1.34) id AA09845; Tue, 26 May 92 10:19:43 -0400
Date: Tue, 26 May 1992 10:19:43 -0400
From: Robert Woody Woodburn <woody@sparta.com>
Message-Id: <9205261419.AA09845@sparta.com>
To: msteenst@bbn.com
Cc: idpr-wg@bbn.com
In-Reply-To: Martha Steenstrup's message of Mon, 25 May 92 21:02:08 -0400 <9205260109.AA05396@sparta.com>
Subject: unavailable virtual gateways

      Unavailable virtual gateways should not have an effect on the routing
      information database.  Rather, as I said in my previous message, a
      route server can use information about unavailable virtual gateways to
      flush routes that traverse such gateways from its route database (i.e.
      route cache).  To use this information, the route server would have to
      keep track of the routes in which each virtual gateway is included,
      and I know that the prototype software doesn't do this.  But I think
      this cache clean up is a function that a route server may well want to
      have.

OK, I can see that.  I change my vote to leaving it in.  It is easy to
compute, and it does make cleaning the route cache easy.

wood y