• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 240
  • Last Modified:

OSPF Creates routing loop between sites

We have just upgraded our core network with some nice new Cisco hardware. The configs from the old have been moved across successfully and have caused no problems thus far.

Our issue appears when we create a static route on Core 1 which points towards Core 2's internet gateway.
Example of static route:

ip route 192.168.100.0 255.255.255.0 172.16.0.254

Instead of Core 2 learning that route it picks up a route pointing back to Core 1 creating a nice routing loop. I was aiming for Core 2 to fire that IP out of its default route (which is 172.16.0.254) where 192.168.100.X resides.

This is the output from sh ip route 192.168.100.1 on Core 2:

Routing entry for 192.168.100.0/24
  Known via "ospf 1", distance 110, metric 20, type extern 2, forward metric 1
  Last update from 192.168.255.253 on Vlan18, 00:00:11 ago
  Routing Descriptor Blocks:
  * 192.168.255.253, from 192.168.255.253, 00:00:11 ago, via Vlan18
      Route metric is 20, traffic share count is 1


The following is the OSPF config of both Core 2, Core 1 had a router-id of 192.168.255.253:

router ospf 1
 router-id 192.168.255.254
 log-adjacency-changes
 redistribute static subnets
 network 10.30.0.0 0.0.1.255 area 0
 network 10.30.30.0 0.0.1.255 area 0
 network 10.32.0.0 0.0.1.255 area 0
 network 10.32.30.0 0.0.1.255 area 0
 network 10.0.0.0 0.255.255.255 area 0
 network 172.16.0.0 0.0.255.255 area 0
 network 192.168.30.0 0.0.0.255 area 0
 network 192.168.31.0 0.0.0.255 area 0
 network 192.168.32.0 0.0.0.255 area 0
 network 192.168.34.0 0.0.1.255 area 0
 network 192.168.36.0 0.0.1.255 area 0
 network 192.168.201.0 0.0.0.255 area 0
 network 192.168.255.0 0.0.0.3 area 0
 network 192.168.255.0 0.0.0.255 area 0
 default-information originate


Any ideas would be greatly appreciated.
0
Jo Cox
Asked:
Jo Cox
1 Solution
 
Craig BeckCommented:
You'll need to add a static to Core2 also.

Core2 has learned that route via Core1 and it doesn't have a more specific route in its own routing table to that subnet, so it will send traffic back to Core1 as that's where it thinks it needs to go to.  The problem is that routing updates don't include the actual gateway address - they include the subnet only and the sending router's interface IP is used as the gateway, so you redistributing static routes isn't doing what you think it's doing.

When a route is received it is basically an announcement from the sending router that it can reach the specified destination.  The interface IP of the sending router is therefore the gateway address for that route.  If a similar route is received from a different router its IP address is also a candidate to route traffic for that subnet and that's how we end up with more than one route sometimes.
0
 
Jo CoxAuthor Commented:
Thanks, that explains it then! A really well explained and easy to understand answer.
Cheers!
0

Featured Post

SMB Security Just Got a Layer Stronger

WatchGuard acquires Percipient Networks to extend protection to the DNS layer, further increasing the value of Total Security Suite.  Learn more about what this means for you and how you can improve your security with WatchGuard today!

Tackle projects and never again get stuck behind a technical roadblock.
Join Now