[c-nsp] BGP sanity check

[Gert Doering]

Hi,

On Sun, Dec 09, 2012 at 05:28:15PM -0500, Chuck Church wrote:
> Thanks.  I guess in my mind the numbers needed to add up more.  Both routers
> are taking a full table, which is more or less the same prefixes with
> different path information.    R2 claims it's sending about 412K to R1, yet
> R1 only sends 100K to R2.  I would think the number should add up
> approximately close to the full table size.  

No.  If both R1 and R2 were to prefer "their" eBGP neighbour, both would
send a full 430K table to each other - and if R1 prefers everything from
R2, it would send 0 prefixes to R2.  

So you'll see something between 430K and 860K if you sum that up.

(Now why R1 or R2 prefer a given prefix via eBGP before iBGP depends
on each individual prefix - path lengths, MED, eBGP-before-iBGP, you
know the BGP decision hierarchy)

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
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 305 bytes
Desc: not available
URL: <https://puck.nether.net/pipermail/cisco-nsp/attachments/20121210/4513f9d9/attachment.sig>