[c-nsp] OSPF clarification

[Alex]

Drew,
If your VLANs are all covered by "network X.Y.Z.W area V" command under 
"router ospf T" then they all are represented as Type-1/Type-2 LSA in OSPF 
LSDB.
And every time a VLAN goes down, LSA-1/2 are reflooded with Age=MaxAge 
(default 3600 sec).
When this VLAn goes back up, LSA-1/2 are recreated and reflooded again.
Every time this happens, all OSPF routers in given area must recalculate 
their SPF, then put OSPF routes in a routing table with new timestamp.
I'd suggest to use OSPF "network" statement for core VLANs only, edge VLANs 
should be injected with "redistrubute connected/static" statement (this 
creates Type-5 LSA for edge VLANs which does not affect SPF).
HTH
Rgds
Alex


--------------------------------------------------
From: "Drew Weaver" <drew.weaver at thenap.com>
Date: 30 November 2009 14:39
To: "Cisco-nsp" <cisco-nsp at puck.nether.net>
Subject: [c-nsp] OSPF clarification

> Hi,
>
> If you have your entire network in a single area (area 0), is it normal 
> that when a VLAN on any switch in that area flaps that all of your routes 
> in OSPF 'reset'?
>
> I don't mean a 'neighbor' VLAN, I mean just any old VLAN with an IP 
> address assigned to it.
>
> Last update from x.x.x.x on Vlan4061, 00:08:42 ago
>
> I mean the 'timers' seem to reset for every single route in OSPF anytime a 
> single VLAN anywhere goes down.
>
> If this is intended, which I am still not sure of, what would be the 
> proper way to 'area' a datacenter?
>
> It doesn't seem like there would be a way to do it so that just the route 
> attached to said VLAN that is flapping would be removed/re-inserted unless 
> you put each VLAN in a separate area.
>
> thanks,
> -Drew
>
>
>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>