Additions to the NSFNET policy-based routing database

Belinda Sue Blair bsb
Thu Apr 9 00:35:42 UTC 1992


I've changed the NWG format to be consistent. It lists the 
number of CONFIGURED networks. It makes no sense to list the
announced networks since this number is constantly changing (duh!).
Thanks for bringing up this point.

-B. Sue Blair
 Merit Network Operations

  >To: Jessica Yu <jyy at>
  >Cc: noc at, nwg at
  >Subject: Re: Additions to the NSFNET policy-based routing database 
  >In-Reply-To: Your message of "Tue, 07 Apr 92 17:21:31 EDT."
  >             <9204072121.AA18405 at> 
  >Date: Tue, 07 Apr 92 18:17:39 -0400
  >From: Erik Sherk <sherk at>
  >> Erik,
  >> The total net numbers CONFIGURED in the NSFNET policy database <> the total
  >> routes actually ADVERTISED to the backbone.  The number that announced 
  >> to nwg every Tue and Fri is the configured number.  In order to get the
  >> actually advertised routes to T1, get on your T1 nss and do a netstat -Nn |wc.
  >> That is the real time advertised number, it may fractuate a bit due to the
  >> tempary outage of particular networks.  But you will get a general idea of
  >> the number.
  >> 						--Jessica
  >	I understand the difference between configured vs. announced networks.
  >That is not what I am asking. I must have been unclear in my previous question,
  >let me try again.
  >	If you grep "Total" from the March 10th network configuration announcment,
  >you get this:
  >aotearoa:[~/Mail/inbox] grep Total inbox/806
  >Total NSFNET T1 announced networks: 4835
  >Total NSFNET T3 announced networks: 2501
  >The total number of networks on each of the backbones has been different
  >in all of the announcements send out by Merit that I have seen. But today,
  >we find that they are the same:
  >aotearoa:[~/Mail/inbox] grep Total inbox/1233
  > Total  T1 announced networks:  5012
  >       Total configured T3 networks:  5012
  >Oh, I see! You have changed the announcment to list "T1 announced" and "T3
  >configured". Would it be too much trouble to ask you to go back to the old

More information about the NANOG mailing list