In short, NO.  You are correct.  I am not looking at the PG right now but it 
probably meant full mesh inside each subAS. 

EBGP between AS is just like regular eBGP...and we know a full mesh cannot 
possibly be supported between thousands of seperate eBGP domains (the 
internet...)


------Original Message------
From: Kim Pedersen
Sender: [email protected]
To: CCIE OSL
Subject: [OSL | CCIE_RS] BGP confederations - Full mesh
Sent: Jul 18, 2009 9:21 AM

Hi All,

  I am doing some practice scenarios, and came across a question 
regarding confederation eBGP peers. My assumption is that based on the 
eBGP nature of confederations, individual "sub-as"'s dont need to be 
fully meshed. But according to the answer to that question, sub-as's 
need a full mesh between them, like normal iBGP. I am not talking about 
within a sub-as, where iBGP rules apply, but simply between these sub-as's.

I have done a simple lab with R1, R2 and R3, being in the same "overall 
AS" 1, and with R4 in AS 777.
R1, R2 and R3 is each in their own sub'as. R1 is connected to R2, which 
is connected to R3. No full mesh. R3 is connected to R4 in regular eBGP 
style, where it is receiving a route. Assuming i have fixed the next-hop 
issues regarding confederations not changing next-hop like regular eBGP, 
i see my route propagated all the way from R4 to R3 to R2 and finally in R1.

  So to conclude, is a full mesh required between sub-as's when using 
confederations?

Sincerely,
Kim Pedersen

-- 

// Freedom Matters
// Follow my progress on: http://kpjungle.wordpress.com

_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com


Regards,

Joe Astorino - CCIE #24347 R&S
Technical Instructor - IPexpert, Inc.
Cell: +1.586.212.6107
Fax: +1.810.454.0130
Mailto:  [email protected]
_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com

Reply via email to