cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
2045
Views
0
Helpful
4
Replies

Multi-link OSPF via MPLS cloud

darren.g
Level 5
Level 5

Hi.

I've recently started work in a new position at a new company, and I'm trying to clean up the mess left by expending too fast and using the wrong people to make said expansion.

Basically, $POE has a bit of a mess regarding connectivity to some client sites, which is done via a third-party MPLS cloud and various links, which I need to sort out.

The remote site(s) in question have two comms links - ethernet over fibre at 4 meg synchronous - for redundancy, along with two routers at each site and several devices which interconnect with client networks spread across the routers - the idea being that if we have a single link/router/device failure, we'll only lose half the connectivity, and continue being able to provide the contracted services.

Unfortunately, the outsourced company who installed the network and setup at these remote sites hasn't enabled both links - one of them is physically connected to the router, however the port is not only unconfigured it's shutdown. The graphis below shows roughly what it looks like

remote-1.jpg

The routers in question at the client site are 2911/K9's, with an SM-ES3-24-P module in them to provide switch ports for device conenction, while the inbuilt router ports are used for the link.

The problem is that the "support" organisation claims that when they enabled the second link and its associated OSPF configuration it caused "a network stability issue with OSPF flapping" - their words, not mine, so they shut it down and left it that way.

Which means that if the active router fails, we lose not only the abiligty to provide our services, we lose the ability to remote configure the second router to enable the redundant link!

I've attached the sanitised configurations from each router, but my thoughts would be that it should be a simple matter of adding another IP address in the /29 range for the "link" subnet and configuring OSPF with a lower priority on the port for the second router to ensure that traffic flows across the primary link, but is able to failover properly should it fail.

Can anyone poke holes in that, or suggest a better way? I'd like to run HSRP or VRRP on the WAN interfaces, but they have no layer 2 connectivity between them, so I don't think I can.

Suggestions? Comments?

Thanks and Cheers.

Darren

1 Accepted Solution

Accepted Solutions

Marwan ALshawi
VIP Alumni
VIP Alumni

If you want to effect the remote router decision about you LAN network then put it on the LAN because ospf calculate the cost on the inbound links

I thinks that's why they shut the interface down they might used

The wan link

HTH

Sent from Cisco Technical Support iPhone App

View solution in original post

4 Replies 4

Marwan ALshawi
VIP Alumni
VIP Alumni

Hi Darren

i believe the issue of falpping happend because once you have the second router OSPF enabled it will start advertise the LAN address 10.50.0.0 and in this case the remote router will have same LAN/network advertised by two next hop going through one link

what you need to do is to increase the OSPF cost PATH through R2 by increasing the OSPF cost of interface GigabitEthernet1/0.2 in R2

for example

interface GigabitEthernet1/0.2

ip ospf priority 0

ip ospf cost 10000

this way the other rotuer 10.100.0.1 will see R1 lower cost and it will be only installed in its routing table while R2 will come in the picture only when R1 stop advertising this network

HTH

if helpful rate

marwanshawi wrote:

Hi Darren

i believe the issue of falpping happend because once you have the second router OSPF enabled it will start advertise the LAN address 10.50.0.0 and in this case the remote router will have same LAN/network advertised by two next hop going through one link

what you need to do is to increase the OSPF cost PATH through R2 by increasing the OSPF cost of interface GigabitEthernet1/0.2 in R2

for example

interface GigabitEthernet1/0.2

ip ospf priority 0

ip ospf cost 10000

this way the other rotuer 10.100.0.1 will see R1 lower cost and it will be only installed in its routing table while R2 will come in the picture only when R1 stop advertising this network

HTH

if helpful rate

Hi.

Would I be best applying that cost to the local network interface, or the WAN network interface? I.E. Should I do it go g1/0.2 or to g0/0 on the second router?

Cheers.

Marwan ALshawi
VIP Alumni
VIP Alumni

If you want to effect the remote router decision about you LAN network then put it on the LAN because ospf calculate the cost on the inbound links

I thinks that's why they shut the interface down they might used

The wan link

HTH

Sent from Cisco Technical Support iPhone App

marwanshawi wrote:

If you want to effect the remote router decision about you LAN network then put it on the LAN because ospf calculate the cost on the inbound links

I thinks that's why they shut the interface down they might used

The wan link

HTH

Sent from Cisco Technical Support iPhone App

Thanks mate - I'll give it a go and see how it works.

Appreciate your help.

Cheers!