[c-nsp] Growing BGP tables

[Rodney Dunn]

On Thu, Jan 27, 2005 at 05:44:25PM +0100, Gert Doering wrote:
> Hi,
> 
> On Thu, Jan 27, 2005 at 11:41:14AM -0500, Rodney Dunn wrote:
> > What about if we just had a per neighbor filter that
> > would filter out more specific prefixes as they come in.
> > Once the more specific is filtered then it's gone until
> > you do a soft clear to get it back.
> > Something along these lines:
> > 
> >  router bgp 100
> >   address-family ipv4 unicast
> >   neighbor x.x.x.x leaked-specifics deny [inbound|outbound]
> 
> This would do fine for me.  
> 
> I wonder how it would work, though, as it might be necessary to revive 
> a more-specific when the supernet is withdrawn?

You would have to do a soft clear to get it back.

To cover for that scenario you could have a default to catch
any routes that are gone.

ie:

more specific is blocked
all less specifics go away

let default take over until a soft clear is done.
This is taking a more memory conservative approach
at the expense of having some overlapping prefixes
be gone from the table.  But that's no different
than a prefix filter blocking all /24's.

It's keeping track of all the parent prefixes that makes
things very difficult to do and scale.

Rodney

> 
> gert
> -- 
> USENET is *not* the non-clickable part of WWW!
>                                                            //www.muc.de/~gert/
> Gert Doering - Munich, Germany                             gert at greenie.muc.de
> fax: +49-89-35655025                        gert at net.informatik.tu-muenchen.de