Cisco Gateway Failover while Network Failure

[youthonprogress]

Hi

I have a sitution where seeking expertise help. I have 2 Gateways and 2 Core Switches connected with Subscriber and Publisher CUCM as shown in the attached pic

VGW1 --> CORE 1 ---> Publisher

VGW2 --> CORE 2 ---> Subscriber

Cores 1 & 2 are HSRP Enabled. and both the Gateways have PRI at each from same provider. And registered as MGCP.

Now if the Core1 fails VGW1 can not see CUCM Cluster and hense the incoming calls are getting silence treatment.

I have 2 interfaces at each VGW routers.

Somehow I want to connect these VGWs with both the cores i.e. G 0/0 to Core 1 and G0/1 to Core 2 of each. Is it possible?

Any help to develop an Idea would be highly appriciable.

Many Thanks!

Nitin

[Jonathan Schulenberg · Accepted solution]

Sure. You should bind the MGCP, SCCP, and other processes to a loopback interface on the router. Then you can choose to build your redundancy at layer two or three:

• With layer two you put both ports in switchport mode and add an SVI for whatever VLAN you're running on them. This is only viable if you have a 6500 VSS or Nexus.
• With layer three you put each interface in a /30 network and extend your routing to them. It is also suggested to increase the weight of one of the links to create an active/standby setup vs. active/active which could result in asymetric routing.

In either scenario you need to tell the core switches about the loopback interface. You could create a static route or extend your dynamic routing protocol of choice to the routers and advertise it.

Please rate helpful replies.