[c-nsp] BGP advertisements more specific than IGP

Scott Granados scott at granados-llc.net
Thu Feb 28 23:57:02 EST 2013

So more specifics are sort of a sledge hammer approach.  If you announce more specifics over one link, assuming a prefix that long is widely readvertised, you'll reroute all the traffic not just have a small effect.  (more specifics always win)  (also it's bad form to not announce unified addressing at your edges)
	Further more, you'll have to have the more specifics in your IGP somewhere to announce and they will have to route meaningfully else you'll have issues reaching internal hosts in these ranges.  If for example you had /24S attached to different internal interfaces you could do a redistribute connected, pass through a route map for appropriate tagging and use that to source the more specifics.
	One thing I bet you're bumping in to is that upstreams or peers of your upstreams probably filter on shorter prefixes than /24 so assuming your IP allocations are from your upstream they will install the larger say /19 that your initial provider is announcing rather than the /24s or even the /23 because they filter on the edges at say /20.  You want to use a looking glass and make sure that you see both announcements equally announced and how widely that's happening.  Routeviews might be a good place to start.  With prefixes that long you may have issues balancing traffic meaningfully.
	Make sure your upstream where you got the allocation is announcing your space independently of the larger allocation.  You might be announcing to your upstream and they might not be passing on that more specific outside their AS.
	Some carriers have communities that will allow you to announce or not to announce to specific peers.  You can try using communities to specifically tag traffic and control how it's announced via the upstreams.   You've got a limited set of options though with prefixes that long. 



On Feb 28, 2013, at 10:11 PM, James Urwiller <jurwiller at americanbb.com> wrote:

> I have a BGP multi-homed invironment that I am having problems balancing inbound traffic, besides prepends which don't seem to be helping anymore, I have heard that announcing my networks more specifically could also influence inbound traffic.  My question is, for example… If I have a /23 that I am using as a /23 in OSPF, can I announce that in BGP more specifically (2, /24's)  without having to them break it up internally as well?  What I foresee happening is this..
> Example:
> BGP:
> Network
> Network
> Network
> I would think in this scenario, the IP addresses and would not have a route in BGP, even though they are valid addresses for use when used as a /23.  Since I would be multi-homed, I would still advertise the network as the aggregate /23 on the circuit I don't want to take as much traffic, so would those IP addresses in this scenario still work, but only through the circuit I advertise as the aggregate??
> James Urwiller
> Network Operations Manager
> CCNA 11567125
> American Broadband
> 402-426-6257 - Office
> 402-278-1875 - Cell
> 402-426-6273 - Fax
> jurwiller at americanbb.com<mailto:jurwiller at americanbb.com>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/

More information about the cisco-nsp mailing list