OSPF cost question

[Bekzod Fakhriddinov]

Hi guys, have a diagram [R1]---Riverbed---[catalystswitchA]---[R2]---[R3]----[catswitchB]--BGP-CE_router-WAN

network 192.168.20.0 is coming from BGP to LAN and ospf cost to it 

from:

Cat_swB metric 0

R3   metric 2 , type 1

R2   metric 3 , type 1

Cat A metric 4, type 1 

R1 metric 4 , type 1

Question is why CAT A and R 1 has the same metric 4 to that route  ?  I can't find it out in the config. 

All interface ospf cost in the path are 1, they are all in the same ospf area. I think R1 metric should be 5 ... 

[Peter Paluch]

Hello Bekzod,

This is interesting - and unexpected :)

It is my understanding that there is no other path R1 could be taking to 192.168.20.0 rather than through CatA. Is this correct?

Also, this should not be possible, but I need to check: Can you verify that R1's interface cost to CatA is at least 1?

Do you think it would be possible to share the full output of the following commands from R1?

show ip ospf database router
show ip ospf database network
show ip ospf database external

Please note that these will essentially allow us to know your topology, your Router IDs, and all your networks so I understand if this is not possible. Still, the answer very likely lies in the contents of LSA1, LSA2, and LSA5, so it is the best approach to check their contents.

Thank you!

Best regards,
Peter

[Bekzod Fakhriddinov · Accepted solution]

 Looks like I found what is it, R1 and R2 also were connected via CAT_A switch  and  were on the same vlan ("l2 tunnel") , that's why R1 was getting ospf update from R2 not from CAT_A .  That's why both Cat and R1 had the same metric 4. Sorry Peter, it was not so interesting as we thought :) . 

thank you