From: Brian McGahan (bmcgahan@internetworkexpert.com)
Date: Thu Jan 20 2005 - 21:07:52 GMT-3
Micah,
The key is that CEF deals with the IP routing table, not the BGP
table. Once the BGP best path has been installed in the IP routing
table the CEF process builds a forwarding table that pairs an IP prefix
with an outgoing interface. So even if you have four full BGP feeds for
a total of 600k+ routes, only 1/4th of these will be best paths
(assuming you're not doing multipath), and therefore only 1/4th will
result in CEF paths.
HTH,
Brian McGahan, CCIE #8593
bmcgahan@internetworkexpert.com
Internetwork Expert, Inc.
http://www.InternetworkExpert.com
Toll Free: 877-224-8987 x 705
Outside US: 775-826-4344 x 705
24/7 Support: http://forum.internetworkexpert.com
Live Chat: http://www.internetworkexpert.com/chat/
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf
Of
> Micah Byers
> Sent: Thursday, January 20, 2005 5:35 PM
> To: mohamed_n@sifycorp.com; peter@cyscoexpert.com
> Cc: ccielab@groupstudy.com
> Subject: RE: Sup2/MSFC2 - Maximum BGP routes per CEF Path
>
> I completely agree with you in that I am sure the 6509 handles that
many
> BGP routes per peer just fine. However, that doesn't really answer
any of
> my questions. You said that CEF Path means the route forwarding done
in
> layer 2, which is really a generic definition that probably more
partially
> describes CEF as a whole.
>
> If you are correct in that definition then you are basically saying
that
> the documentation for all of the IOS releases for this module are
wrong
> (which could be the case... but not that likely) because otherwise it
> wouldn't handle any more than 150,00 BGP routes.
>
> I am guessing nobody on this list has ran across this, or knows
exactly
> what the author of the document meant by the table presented (judging
by
> the lack of response). And that's ok. I was just trying to see if I
> could get the specifics... Thanks for the reply!
>
>
> -Micah-
>
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf
Of
> mohamed_n@sifycorp.com
> Sent: Thursday, January 20, 2005 1:43 AM
> To: peter@cyscoexpert.com
> Cc: ccielab@groupstudy.com
> Subject: Re: Sup2/MSFC2 - Maximum BGP routes per CEF Path
>
> Iam sure it can accomadate 250k + routes,i have tested!
> CEF path means the route forwarding done in Layer2.The packet
forwarding
> done hit the route processor.So speed is high..And SUP720 is still
more a
> gaint than SUP2.
>
> Thanks
> Mohamed.
>
>
> peter@cyscoexpert.com:
>
> > Micah,
> >
> > They probably mean hardware routed/switched path, since this is what
> > CEF
> > does.
> > With the new Sup720-3BXL for Cat6500 this is increased to 1 million
> > routes.
> >
> > Peter
> > #7247 (R&S, Security, SP)
> > CXP Corp.
> > 4433 W. Touhy Ave. Suite 410
> > Lincolnwood, IL 60712
> > Phone (847) 674-3392
> > Fax (847) 674-2625
> >
> > ----- Original Message -----
> > From: "Micah Byers" <mbyers@gramtel.net>
> > To: <ccielab@groupstudy.com>
> > Sent: Wednesday, January 19, 2005 10:15 AM
> > Subject: Sup2/MSFC2 - Maximum BGP routes per CEF Path
> >
> >
> > >
> >
>
http://www.cisco.com/univercd/cc/td/doc/product/lan/cat6000/relnotes/ol_
39
> 75
> > .htm
> > >
> > >
> > >
> > > The link above states that the Sup2/MSFC2 with 512MB DRAM can
sustain
> > a
> > maximum of 150,000 BGP routes per CEF Path. What exactly does this
> > mean?
> > Has anyone ran into this? The Internet BGP table with full routes
is
> > over
> > 150,000 routes now so if this was an issue of routes per peer I
would
> > think
> > this would have become a big problem.
> > >
> > >
> >
This archive was generated by hypermail 2.1.4 : Wed Feb 02 2005 - 22:10:24 GMT-3