From: Jason Sinclair (sinclairj@xxxxxxxxxxxxxxx)
Date: Tue Aug 27 2002 - 01:16:12 GMT-3
Omer,
I think Peter van Oene hit on an excellent point about the problem with
certifications. In the lab you would need to configure your scenario exactly
as they ask. However, in real life you would certainly configure (to use
your term) MPOR as there is no sane reason to design a network with no
redundancy if the links are already there to support it. So, if you are
asking this from a real world perspective, the answer is definitely use
MPOR, however in the lab make sure you read what they require. I am pretty
sure they would be explicit in their questions, and if not ask the proctor.
Regards,
Jason Sinclair CCIE #9100
Manager, Network Control Centre
POWERTEL
55 Clarence Street,
SYDNEY NSW 2000
AUSTRALIA
office: + 61 2 8264 3820
mobile: + 61 416 105 858
email: sinclairj@powertel.com.au
-----Original Message-----
From: Omer Ansari [mailto:omer@ansari.com]
Sent: Tuesday, 27 August 2002 09:38
To: ccielab@groupstudy.com
Subject: design question: multiple points of redistribution
All,
When doing different lab scenarios, I've noticed they dont explicitly
state if you should have multiple points of redistribution. (MPOR)
My take on when MPOR is they are only needed when you are asked to
implement redundancy in your network.
thus in the following scenario e.g.
---{BGP}---
| |
(r1) (r2)
| |
ospf ospf
|----r3---|
would it be wise to redist bgp into ospf on both r1 and r2, even if
there is NO requirement in the lab that r3 should maintain connectivity to
bgp networks if one of its two upstream links croak?
regards,
Omer
This archive was generated by hypermail 2.1.4 : Sat Sep 07 2002 - 19:48:39 GMT-3