Overview

Virtual Router Redundancy Protocol (VRRP) eliminates a single point of failure by allowing two or more physical routers to share a virtual IP and MAC address. This FAQ answers critical pre-sales and post-sales questions about optical module compatibility, throughput, backplane capacity, and step-by-step CLI configuration for enterprise and data center deployments.

Frequently Asked Questions

Q1: What is the maximum switching throughput and backplane capacity for VRRP-enabled routers?

The maximum switching throughput depends on the hardware platform; typical carrier-grade VRRP-capable routers range from 480 Gbps to 25.6 Tbps with backplane capacities exceeding 12 Tbps for modular chassis. This line-rate performance remains active even during VRRP failover because the protocol operates at the control plane while data plane forwarding continues via hardware ASICs.

Q2: Which optical modules (SFP, SFP+, QSFP28) are compatible with VRRP deployment?

VRRP is media-agnostic and supports any standard optical module—1G SFP, 10G SFP+, 25G SFP28, 40G QSFP+, and 100G QSFP28—provided the router's physical interface is compatible. For multivendor VRRP groups, ensure both routers use identical transceiver types on the same VLAN to avoid link asymmetry during failover.

Q3: What is the basic CLI configuration to set up VRRP on two routers?

First, assign the same virtual IP address (e.g., 192.168.1.1) on both routers using 'vrrp 1 ip 192.168.1.1'. Set priority (e.g., 'vrrp 1 priority 150' on master, 'priority 100' on backup). Configure preemption ('vrrp 1 preempt') and advertisement interval ('vrrp 1 timers advertise 1'). Always bind VRRP to a physical or VLAN interface and verify with 'show vrrp brief'.

Q4: How does VRRP differ from previous router redundancy protocols like HSRP or GLBP?

VRRP (IEEE 802.1Q standard) differs from Cisco-proprietary HSRP by using a single virtual IP and MAC address without separate 'standby' IP addressing; GLBP adds load balancing across multiple routers while VRRP natively supports active/backup only. VRRPv3 adds IPv6 support, and unlike HSRP, VRRP allows using the real interface IP of the master as the virtual IP.

Q5: How do I configure stacking interfaces or virtual chassis to work with VRRP?

Stacking interfaces (e.g., StackWise, VPC, or MLAG) act as a single logical node; VRRP should be configured on the stack's VLAN interfaces, not on individual stack ports. For multi-chassis LAG (MC-LAG) stacks, run VRRP on the logical SVI (Switch Virtual Interface) and ensure the stack split detection is disabled to prevent VRRP split-brain scenarios.

Q6: What is the correct troubleshooting procedure for redundant power supply failover during VRRP events?

If a power supply fails on the VRRP master, the master may reboot or stall—causing VRRP advertisement timeouts (default 1 second) and triggering failover to the backup. Troubleshoot by checking 'show power' for PSU status, verifying VRRP 'track' object (e.g., 'track 1 interface gig0/1 line-protocol') to degrade priority upon power loss, and testing with 'redundancy force-switchover' to simulate failover without unplugging.

Q7: What is the maximum throughput in Gbps supported across a VRRP failover domain?

VRRP itself introduces negligible forwarding overhead (<0.1% CPU), so the maximum aggregate throughput equals the sum of the active router's interface capacity—commonly 40 Gbps, 100 Gbps, or 400 Gbps per uplink. During failover, subsecond convergence (typically 3-5 seconds using default timers) occurs without throughput loss if both routers share identical forwarding tables via OSPF or BGP.

Q8: How do I test and validate VRRP redundancy for a live production network?

Validate by performing three tests: (1) graceful failover with 'clear vrrp 1 force-switchover', (2) power failure by disconnecting the master's AC feed, and (3) link failure by shutting the master's tracked uplink. Use continuous ping to the virtual IP with 1ms intervals; expect 1-3 lost pings during failover. Verify with 'debug vrrp packets' and syslog for 'backup to master' state changes.