CSS11051 Interface Redundancy

[tweber]

It appears that there are a couple of ways to do redundancy - box to box using a separate interface for heartbeat, etc. or VIP and Interface redundancy using VRRP.

I am leaning more towards using VIP and Interface redundancy since I believe there is more control (by using priority and preempt the master should always be known).

I will have 2 CSSes, 2 SCAs, 2 firewalls, and multiple web servers attached to 2 L2 switches that will be trunked together. A separate VLAN will be created for the CSSes - SCAs, CSSes - Firewall, and CSSes - Web servers.

If the CSS fails VRRP should switch to the back CSS, if the primary L2 switch fails VRRP should switch to the back CSS.

Finally to the question (sorry it took a while):

What happens if just one interface fails on the CSS?

I need to have a setup so that if any interface fails all failover to the backup CSS.

What is the best way to do this?

Thanks,

Tim

[#2]

Tim, To detect links state, you can try the redundancy-phy command can make redundancy much more sensitive to a failed link. Be careful using it and make sure you do NOT use it on the redundancy-protocol link between the two CSS's. This will cause the two CSS's to flap back and forth.

Jim