cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
4546
Views
5
Helpful
7
Replies

Multiple OSPF routing process same networks statements

fabrice75
Level 1
Level 1

Hi,

I am migrating a current OSPF network  to a new one so I have created an new OSPF process on all devices in the network but with a larger administrative distance so that I don't disturb the exisiting one. The plan was to do the migration by further increasing the admin distance on the old OSPF process and then remove it at a later date.

The OSPF peering addressing for the new process is unique to that process so no issues there but when I copy all the networks statements from the old OSPF process to the new one they do not appear on "show ip OSPF interface brief"

In fact I realised that I have to remove them first from the old OSPF process then put it on the new OSPF process otherwise it won't work.

I want to be clear that the network statements that I want to copy are not for neighbor purpose (they are passive interfaces) the purpose is just to have the network announced in both OSPF processes so I have almost the exact same OSPF database on both. In other words I know that it is not possible to run OSPF neighbor over the same IPs on from two differents processes on the same box but I was expecting to see those network present on both OSPF databases which is not the case.

 

Thanks,

 

 

1 Accepted Solution

Accepted Solutions

Yes this is normal behavior. If you want both processes to advertise these networks then I suggest that you try using redistribute connected which should advertise the same networks in both processes. If you want some but not all networks advertised (as seems to be the case in your example) then you can use a route map to control the redistribution and permit only the networks that you do want advertised.

 

HTH

 

Rick

HTH

Rick

View solution in original post

7 Replies 7

LA-Engineer
Level 1
Level 1

So you're planning on running multiple OSPF processes on the same interface?  I don't think that is supported.  OSPF doesn't have a mechanism to signal to each other what its process ID is, unlike EIGRP which advertises its AS in its hellos.

Hi Guys,

Thanks for your answers, however I fear that I have not been understood.

Agreed with you LA, I am not trying to establish neighbor for two OSPF process on same interface, this is not supported.

I can't put production config here but I have writen a sample configuration. Please note that it is not applied to any devices so won't be able to produce any output commands this is just to give you a better idea.

 

Router A
!
interface Loopback0
 description MGNT Interface
 ip address 1.1.1.1 255.255.255.255
!
interface GigabitEthernet1/1
  description User LAN
  ip address 192.168.1.1 255.255.255.0 

interface GigabitEthernet1/10
  description To router B fo OSPF 10 ONLY
  ip address 10.1.1.1 255.255.255.252
!
interface GigabitEthernet1/20
  description To router B fo OSPF 20 ONLY
  ip address 20.1.1.1 255.255.255.252
!
router ospf 10 
 router-id 10.1.1.1
 log-adjacency-changes
 passive-interface default
 no passive-interface GigabitEthernet1/10
 network 10.1.1.0 0.0.0.3 area 0
 network 1.1.1.1 0.0.0.0 area 0
 network 192.168.1.0 0.0.0.255 are 0
!
router ospf 20 
 router-id 20.1.1.1
 log-adjacency-changes
 passive-interface default
 no passive-interface GigabitEthernet1/20
 network 20.1.1.0 0.0.0.3 area 0
 network 1.1.1.1 0.0.0.0 area 0
 network 192.168.1.0 0.0.0.255 are 0
 distance 150


Router B
!
interface GigabitEthernet1/10
  description To router A fo OSPF 10 ONLY
  ip address 10.1.1.2 255.255.255.252
!
interface GigabitEthernet1/20
  description To router A fo OSPF 20 ONLY
  ip address 20.1.1.2 255.255.255.252
!
router ospf 10 
 router-id 10.1.1.2
 log-adjacency-changes
 passive-interface default
 no passive-interface GigabitEthernet1/10
 network 10.1.1.0 0.0.0.3 area 0
!
router ospf 20 
 router-id 20.1.1.2
 log-adjacency-changes
 passive-interface default
 no passive-interface GigabitEthernet1/20
 network 20.1.1.0 0.0.0.3 area 0
 distance 150

 

Now if you look on Router B you will not see network 1.1.1.1 & 192.168.1.0 under the OSPF process 20 database if you have entered the network command on OSPF 10 first on Router A.

If you look on Router A with "show ip ospf int brief" you won't even see interface Loo0 and Gi1/1 listed.

After tresting around it seems that I can't have both of those network present in both database, whatever I try.

If I want have those networks present in OSPF 20 I have first to remove them from OSPF 10 and then put them on OSPF 20.

I just wanted to know if this is normal behaviour or not?

 

Thanks,

 

 

 

 

Yes this is normal behavior. If you want both processes to advertise these networks then I suggest that you try using redistribute connected which should advertise the same networks in both processes. If you want some but not all networks advertised (as seems to be the case in your example) then you can use a route map to control the redistribution and permit only the networks that you do want advertised.

 

HTH

 

Rick

HTH

Rick

Thanks,

It would be interesting to know why is that?

I do understand that you would not want to do that for neighboring but for simple network annoucement, I don't see an issue with it really.

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

It's probably because some Cisco "designer/architect" decided an interface could only be "owned" by one process.  (Somewhat similar how Cisco devices won't normally allow two interfaces within the same subnet.)  This might have been to avoid accidental overlaps.  As Rick notes you can still redistribute.  (If fact, I wonder what happens if you try redistributing connected to both OSPF processes.)

Parvesh Paliwal
Level 3
Level 3

Another workaround can be using VRFs.

 

An interface in OSPF can belong to multiple process id but remember a process id is local to the router and need to workout, how you are planning to sync these routers.

Parvesh Paliwal
Level 3
Level 3

The document may help out:

http://www.cisco.com/c/en/us/support/docs/ip/open-shortest-path-first-ospf/4170-ospfprocesses.html

 

Also, share the config you are trying to apply.

 

Hope it helps out.

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:

Innovations in Cisco Full Stack Observability - A new webinar from Cisco