OSPF coop site

chipsch21 · ‎11-08-2010

Have a rough idea of how to go about this but not 100% positive yet.

So we have a multi-area OSPF environment currently. We have been

tasked with adding a coop site into the mix that is physically

located off one of the ABR's. This coop is obviously for failover should

something happen to the current area 0. That being said I would think

the coop would need to be area 0 also in order to keep the communication

going in the event of failure at the original area 0. In order to do this I

was thinking maybe a virtual-link to connect the area 0's so that the

coop area 0 would still receive all required routing updates. I imagine

their would also have to be some routing policies put into place also

to keep traffic from traversing the coop area 0 unless a failover has occurred.

Am I on the right path or is my mind way out in left field.

Giuseppe Larosa · ‎11-13-2010

Hello Chipsch21,

>> In order to do this I

was thinking maybe a virtual-link to connect the area 0's so that the

coop area 0 would still receive all required routing updates.

Virtual links are not recommended as a permanent solution, a possible solution can be to just increase OSPF cost on links to/from this coop site so that it is bypassed by traffic in normal network scenario even if it is part of area 0.

In case of failure traffic starts to use the high cost links

May I ask you what do you mean with coop site?

I tried to think of a sort of disaster recovery site.

Hope to help

Giuseppe

Richard Burts · ‎11-13-2010

I generally hear coop in terms of Continuity of Operations, and that would make sense in this context.

to the original poster:

You describe the coop as connected through one of the ABR routers. In this case the ABR already has connection to the existing area 0. If you configure an interface for connection to the coop site and assign it to area 0, then everything should work and there is no need of virtual links. If you want to control traffic and to make it unlikely that traffic would flow to the coop while the primary was active I would think that a high cost assigned to the OSPF interface to the coop would be sufficient to accomplish what you want.

HTH

Rick

HTH

Rick

aqeel.karim · ‎11-15-2010

I agree with Richard, changing the OSPF cost on the primary and Coop site would accomplish what you need when your Prim site fails. This could be good a bad... since most prim sites don't completely fail, but there are constant network outages. Hopefully, your COOP site is a Hot site and not anything less.

Reason, if you don't control your network outage a forget that you are failing over to the COOP, all your traffic would go to the site that may not have the exact set of resources as the Primary location, creating a dead-end for your network users. In this position, your organization may need to take a closer look at your COOP site planning.

But from a simple routing perspective Richard called it exactly and appears to be the simpliest method.