On Wed, Dec 2, 2009 at 20:23, Hoogen <hoogen82_at_gmail.com> wrote:
> This wasn't for real design :) This was for my studies.. the "what if"
> questions that keep popping up in my head.. Cause I would assume the exam
> would probably want us to know most options..
> Mark when you say bad design choice.. do you mean disabling the route
> reflection or having no peering between R3 and R4... ?? I would assume that
> having no peering between R3 and R4 is not a bad choice since there is no
> BGP routes installed on them that needs to be passed on..Or is this a
> mandatory desing requirement that when reflection is disabled we need to
> have full mesh peering.
When I say bad design, I mean the combination of the two things in
this particular case. Yes, you do not have any routes right now and it
looks like a good idea. However, in real networks, things change.
Configuring no reflection and no peering could cause you gruesome
troubleshooting in the middle of the night, years from now, when ideas
fade and paper turns pale yellow.
For learning - this is a fine scenario. If there are no routes to be
exchanged, there is no point in having peering between them. The will
learn what they need from their RR and RR will let non-clients what
they need to know.
P.S. I'll allow it once, second time you get a warning - there is an
"o" at the end ;-).
-- Marko Milivojevic - CCIE #18427 Senior Technical Instructor - IPexpert Mailto: markom_at_ipexpert.com Telephone: +1.810.326.1444 Live Assistance, Please visit: http://www.ipexpert.com/chat eFax: +1.810.454.0130 Blogs and organic groups at http://www.ccie.netReceived on Wed Dec 02 2009 - 20:28:04 ART
This archive was generated by hypermail 2.2.0 : Sat Jan 02 2010 - 11:11:07 ART