Weird ospf topologies

From: ccie2be (ccie2be@nyc.rr.com)
Date: Wed Apr 27 2005 - 14:46:29 GMT-3


Hi guys,
 
I'm trying to collect all the weird, non-standard ospf topologies that are
never used in the real world but that Cisco loves to come up with in the
lab.
 
For example, consider this topology:
 
area 1 rtr-1 area 2 rtr-2 area 3
 
At first, this looks like an illegal topology because there's no area 0.
However, if a virtual link is configured between rtr-1 and rtr-2, all is
well.
 
Here's another one.
 
area 0 rtr-1 area 2 rtr-2 area 3 rtr-3 area 4
 
At first, this looks also illegal but by adding a 2nd virtual-link between
rtr-2 and rtr-3, this will also work since it's OK to chain v-links
together. (BTW, off-hand, I don't know if area 0 were changed to area 1 if
this would still work but I imagine it would. Any comments?
 
So, this is a call for all weird, non-standard ospf topologies.
 
Be creative. Try to think of anything outside of the norm because you know
Cisco is doing the same.
 
Don't limit your responses to just things having to do with virtual-links
either. If you know of or can think of a non-intuitive way for a certain
ospf requirement to be worded, please share your thoughts whether it's about
summarization, network type, a seldom used ospf feature, nssa areas, or
anything else regarding ospf.
 
 
I know that Cisco can find other things to trip us up on during the lab but
let's at least remove ospf as a potential pitfall.
 
Tim



This archive was generated by hypermail 2.1.4 : Tue May 03 2005 - 07:55:09 GMT-3