In my last post, tried to explain OSPF High Availability with SSO,NSF and NSR. In this post I am trying to give a snapshot how does OSPF behaves with SSO,NSF and NSR.
OSPF without SSO/NSF
1. All the OSPF neighborship are up and running. At any point of time, Router A has some issue with primary RP and it gets restarted.
2. Adjacency between router A and B will down.
3. Router B will remove the adjacency and clear the entire forwarding table for router A.
4. Router B will inform the change about router A to all its connected peers named C,D and E.
5. Now router A is up again after reboot and establish a peer relationship with router B.
6. Router B will add the routes in its routing table.
7. Router B will inform the change to its peer neighbors.
OSPF with SSO/NSF
1. All the OSPF neighborship are up and running and Graceful Restart Capability are exchanged between Router A and Router B.
2. At any point of time, Router A has some issue with primary RP and switch to the secondary RP.
3. Router B doesn’t remove the associated routes from its routing table.
4. Router B doesn’t inform its peer about the change.
5. Router A standby RP re-establishes adjacency.
6. Router B updates router A with its routing information.
7. Router A updates router B with its routing information.
OSPF with NSR (Non Stop Routing)
1. Router A synchronizes it’s all OSPF states and databases to the secondary or standby RP.
2. Router A primary RP fails.
3. Router B doesn’t remove routes from its table and doesn’t inform the neighbors about the OSPF change.
4. Router A standby RP continues forwarding while the using the states exchange by the primary RP.