Redistributing Loopbacks into OSPF area 0

Redistributing loopbacks into Area 0 of OSPF looks to be a common cause of creating a discontiguous area 0 if a link fails.

For example task 3.1 in Vol 2 Lab3 - hub and spoke FR with loopbacks advertised into OSPF area 0. R5 is the hub and R1, R2 and R4 the spokes.

If the link between R1 and R5 or R2 and R5 fails then the loopbacks at either R1 or R2 become disconnected from the OSPF area 0 and a discontiguous area 0 is created.

I accept I would need a virtual link if R2 had say Area 22 behind it.

However, do I need to protect the R1 and R2 LOOPBACKS from possible disconnection from area 0 with virtual links in the lab exam?



  • Hey Huxley,

    I may well have mis-understood your question here, but the way I'd look at that particular lab, if say R1 becomes disconnected from R5, then it disconnected from the whole OSPF domain anyway, so area's don't really matter. It's now a router connected only via EIGRP.

    As far as the lab's concerned - have they asked you to worry about what happens in this link fails? If not then no worries. If yes, then to maintain connectivity to it you could either put in a tunnel (which would maintain the OSPF integrity), or do something clever with the redistribution into/outof EIGRP (which for those who havn't seen the topopology is what runs from the back of R1 to the next EIGRP/OSPF border router). 

    If there was OSPF area 12 at the back instead of EIGRP, then there would be no problem, as redistributing the connected interface into OSPF will advertise T5 LSA's into ALL areas, so when the link is up, it will be seen in area 0, and when it goes down, it'll be forwarded through R2 back into A0.

    Again sorry if I've mi-understood what you're looking for here..



Sign In or Register to comment.