Home page logo

nanog logo nanog mailing list archives

Re: BGP4 peering wiht Sprint, and MCI
From: Curtis Villamizar <curtis () ans net>
Date: Mon, 16 Oct 1995 11:35:32 -0400

In message <Pine.LNX.3.91.951014142731.10170A-100000 () netrail net>, Nathan Strat
ton writes:

We have a 486 DX4 100 with 40 megs of ram running FreeBSD and gated with a 
T1 to Sprint and a T1 to MCI. I have a few questions that someone may be 
able to answer. 

If my Sprint T1 goes down, how long will it take for route to drop out of 
the routing table?

If the interface is marked down, worst case is 15 seconds for gated to
scan the interfaces.  Since most interfaces don't get marked down when
signal is lost, you are likely to end up waiting the BGP 180 second
timeout (assuming you are running BGP).  You can set BGP to send
keepalives more often, though this can be a bad idea for other
reasons.  The holdtime is always 3 time the keepalive interval.

I have peering going with sprint and MCI and I have about 31,000 routes 
in my routing table. When I called up sprint and MCI they said they were 
not getting routes form me. How do I advertize my CIDR blocks to sprint 
and mci via BGP4?

I have:

near the top:

  aggregate masklen 24 {
    proto <your-igp> { masklen 24 refines;   # assuming you have only subnets
  aggregate masklen 19 {
    proto <your-igp> { masklen 19 refines;   # accept any subnmet into the aggr

near the bottom:

  export proto bgp 3651 {
    proto aggregate { masklen 24; masklen 19; masklen 18;

Nathan Stratton                 CEO, NetRail, Inc.    Your Gateway to the Wor
Phone   (703)524-4800                        NetRail, Inc.
Fax     (703)534-5033                          2007 N. 15 St. Suite B-5
Email   sales () netrail net                      Arlington, Va. 22201
WWW     http://www.netrail.net/                Access: (703) 524-4802 guest

  By Date           By Thread  

Current thread:
[ Nmap | Sec Tools | Mailing Lists | Site News | About/Contact | Advertising | Privacy ]