cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
938
Views
0
Helpful
2
Replies

OSPF Type 5 LSAs for default route

Lukas Auer
Level 4
Level 4

I have two data centers interconnected through a MPLS backbone running MP-BGP (typical L3 VPN setup with two sites).

If rz1n1-dc redistributes its locally configured static default route into OSPF, the following happens:
- one type 5 LSA for 0.0.0.0 in the left OSPF domain with Advertising Router: 10.40.32.246 (=rz1n1-dc)
- two type 5 LSAs for 0.0.0.0 in the right OSPF domain with Advertising Router: 10.40.31.251 (=rz2n1-c) and 10.40.31.252 (=rz2n2-c)

That's pretty much what I would expect to happen. The right side learns the default route through BGP and the two PE routers rz2n1-c and rz2n2-c redistribute it into OSPF.

 

Now comes the interesting part:
If rz2n1-dc also starts to redistribute its locally configured static default route into OSPF, the following happens (so we now have rz1n1-dc and rz2n1-dc redistributing a default route into OSPF):
- one type 5 LSA for 0.0.0.0 in the left OSPF domain with Advertising Router: 10.40.32.246 (=rz1n1-dc)
- two type 5 LSAs for 0.0.0.0 in the right OSPF domain with...
  > Advertising router: 10.40.31.246 (rz2n1-dc)  (the router which redistributes its static default route into OSPF) (expected)
  > Advertising router: 10.40.31.252 (rz2n1-c) (one of the two PE routers which redistributes the default route from the left side from BGP to OSPF)  (not expected)

I don't understand why I see the second type 5 LSA on the right side. I would expect to see only one type 5 LSA from rz2n1-dc (just like I see only one type 5 LSA on the left side).
By the way: If I look at the OSPF database for a different route (not the default route) which is configured identically (also redistributed from static into OSPF on rz1n1-dc and rz2n1-dc), I only see one type 5 LSA on both sides, just like I would expect.

Any ideas what I am missing? Is this some expected OSPF feature related to default route I am not aware of?
By the way: I see the same behavior in production environment (running NX-OS) and also in GNS3 simulation (running IOS).

rz2n1-dc#sh ip ospf database external 0.0.0.0

            OSPF Router with ID (10.40.31.246) (Process ID 1)

                Type-5 AS External Link States

  LS age: 11
  Options: (No TOS-capability, DC)
  LS Type: AS External Link
  Link State ID: 0.0.0.0 (External Network Number )
  Advertising Router: 10.40.31.246
  LS Seq Number: 80000001
  Checksum: 0x93C9
  Length: 36
  Network Mask: /0
        Metric Type: 2 (Larger than any link state path)
        TOS: 0
        Metric: 1
        Forward Address: 10.0.1.2
        External Route Tag: 1

  Routing Bit Set on this LSA
  LS age: 215
  Options: (No TOS-capability, DC)
  LS Type: AS External Link
  Link State ID: 0.0.0.0 (External Network Number )
  Advertising Router: 10.40.31.252
  LS Seq Number: 80000022
  Checksum: 0xE181
  Length: 36
  Network Mask: /0
        Metric Type: 2 (Larger than any link state path)
        TOS: 0
        Metric: 1
        Forward Address: 0.0.0.0
        External Route Tag: 3489725969

 

 

2 Replies 2

MOHAN MUDUNURI
Level 1
Level 1

Hi Lukas

 

Can you let me know what and where this 10.0.1.2 is configured in your network. If possible running config of rz2ni-dc as well

 

Regards

Mohan

Hi Mohan,

thanks for your answer. We recently solved the problem by reloading the N7k switch and haven't seen the issue again since then.

Lukas

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:

Review Cisco Networking products for a $25 gift card