![]() ![]() We’ve been using dual Supervisors in them for IP Telephony readiness or IPT-using sites. I started thinking about this in the context of wiring closet switches with dual uplinks to the building distribution layer. NSF clearly wins if there is no other routing alternative. And we’ve seen or heard about Cisco doing demos where a ping was running through a switch, failover triggered from the CLI, and nary a ping lost. If the neighbors don’t hear from the spare Supervisor within a time window (“I’m here, I’m alive”), the adjacencies bounce - protection against a completely failed / de-powered chassis.The switch’s neighbors perform Graceful Restart to restore adjacency to the new Supervisor (same address, different brain) and synch up, rather than bouncing the routing adjacency and restarting things.The spare Supervisor keeps links up (except of course for those on a failed Supervisor module blade).When a Supervisor fails, several things happen: They do this by keeping L2 and 元 state synchronized between the Supervisors. Non-Stop Forwarding (NSF) and Stateful Switchover (SSO) are designed so a dual-Supervisor 6500 (or other device) can minimize packet loss during a Supervisor failure. How they interact, good but apparently a little incomplete. The explanations of each topic are pretty good. ![]() To some extent, the more I read, the more puzzled I got. My preference would be to do it from core to the edge.I’ve been reading a lot of documents, looking for clarity about Non-Stop Forwarding / High Availability / Resiliency and how it interacts with Fast Re-Routing. You could certainly do the conversion from the edge to the core or from the core to the edge. You can do that by changing the AD of EIGRP to be lower than OSPF or by removing EIGRP from the router configurations. After you have verified that OSPF is performing as expected then you can switch over from EIGRP to OSPF. So implementing OSPF will not have much impact on the running network, and you will have opportunity to verify that OSPF is running as expected, that OSPF adjacencies are formed as expected, that entries are created in the OSPF data base as expected. One interesting aspect of converting a running network from EIGRP to OSPF is that as you implement OSPF it will run in parallel with EIGRP and the EIGRP routes will be preferred (based on the AD of the protocols). So election of DR/BDR is not much of a concern in converting the network from EIGRP to OSPF. A network segment at the edge would have its own DR/BDR, a network segment in the middle wold have its own DR/BDR, and the network segment at the core would have its own DR/BDR. So most OSPF implementations will have multiple DR. As was pointed out in a previous response the OSPF DR/BDR are elected per multi access segment in the network. The original post exhibits a belief that is common when people are learning about OSPF that there would be a single DR for the OSPF domain. How important is it to actually have one of your routers be elected as BDR in OSPF? My only worry is that if I set all of the exterior routers with a priority of 0 will they still form neighbor relationships with each other and then cause redundant traffic to be passed around the network until I configure the core due to the fact that there is no designated router?Īdditionally, if I go this way then there will never be a backup designated router. My current idea is that I figure all exterior routers with an OSPF priority of 0 then once I get to the core configure that with an OSPF priority of 2 then the routes will be learned and the core will be the designated router. I know that once I get to the core I can start shutting interfaces and restarting the OSPF process to force the core to be the DR but I really do not want to go that route. I am worried that if I start enabling OSPF on the exterior first then another router would become the designated router before I get to the core, even if I configure the core with a higher OSPF priority due to the fact that an OSPF DR has already been elected. That being said my end goal is that the core be the designated router. My thought is that I would start configuring OSPF on the exterior and work my way to the core. I am tasked with migrating a network from EIGRP to OSPF. ![]()
0 Comments
Leave a Reply. |