What Is Failover?

Failover is the process of switching from a failed or unavailable primary component to an alternative component or path.

The basic idea is:

         PRIMARY PATH
              |
              X  ← Failure
              |
              ↓
         FAILOVER
              |
              ↓
      ALTERNATIVE PATH
              |
              ↓
          Network
    

For example, imagine a company has two Internet connections. If the primary Internet connection fails, the network can use the secondary connection.

                Internet
               /        \
            ISP 1       ISP 2
           PRIMARY    SECONDARY
              |           |
              |           |
           Router 1    Router 2
              \           /
               \         /
                LAN
                 |
               Users

         ISP 1 fails
              ↓
         Traffic uses
         ISP 2
    

The exact failover mechanism depends on the network design. Routing protocols, tracking mechanisms, gateway redundancy protocols, or SD-WAN technologies can be used for different scenarios.

What Are Redundant Links?

A redundant link is an additional network connection that provides an alternative path if another link fails.

         Switch A
         +-------+
         |       |
         +---+---+
             |
      +------+------+
      |             |
   Link 1         Link 2
      |             |
      +------+------+
             |
         Switch B
         +-------+
    

If Link 1 fails, Link 2 can potentially provide connectivity. However, connecting switches with multiple independent Layer 2 links can create loops. Protocols such as STP/RSTP or technologies such as EtherChannel are therefore important in appropriate designs.

Redundant Links vs EtherChannel

There are different ways to design multiple connections between network devices.

EtherChannel using LACP can allow multiple physical links to operate as one logical Port-Channel. If one member link fails, remaining members can continue to carry traffic, depending on the configuration.

What Is High Availability?

High availability (HA) is a design approach intended to keep systems and services available despite component failures or maintenance events.

      NORMAL OPERATION

    +-----------+       +-----------+
    | Device A  |       | Device B  |
    | Available |       | Available |
    +-----------+       +-----------+
          \                 /
           \               /
            +-------------+
                  LAN


      FAILURE

    +-----------+       +-----------+
    | Device A  |       | Device B  |
    |   FAILED  |       | Available |
    +-----------+       +-----------+
                            |
                          Traffic
    

High availability is broader than simply having two devices. It can involve redundant hardware, links, power, routing paths, gateway mechanisms, monitoring, failover procedures, and operational processes.

High Availability vs Redundancy

These terms are closely related but are not exactly the same.

What Is HSRP?

HSRP stands for Hot Standby Router Protocol.

HSRP is a Cisco first-hop redundancy protocol that allows multiple routers to work together to provide a resilient default gateway for hosts.

                LAN
                 |
         Virtual Gateway
          192.168.10.1
                 |
          +------+------+
          |             |
       Router 1      Router 2
        ACTIVE        STANDBY
          |             |
       Gateway       Backup
    

Instead of configuring hosts to depend directly on one physical router as their default gateway, the hosts can use a virtual IP address.

If the active HSRP router becomes unavailable, another router participating in the HSRP group can assume the active role.

What Is VRRP?

VRRP stands for Virtual Router Redundancy Protocol.

VRRP provides a standards-based mechanism for router or first-hop redundancy. Multiple routers can participate in a virtual router arrangement so that hosts can continue using a virtual default gateway if the active device fails.

                 LAN
                  |
          Virtual Gateway
                  |
         +--------+--------+
         |                 |
      Router A          Router B
       Master            Backup
    

Terminology and configuration details differ between HSRP and VRRP, but both address the general problem of providing a resilient first-hop gateway.

HSRP vs VRRP

Why Do We Need a Virtual Gateway?

Consider a network where every computer uses Router 1's physical IP address as its default gateway.

```

PC 1 ----
PC 2 -----+---- Router 1
PC 3 ----/       |
|
Internet 

If Router 1 fails, the computers may lose their path to external networks even if Router 2 is available.

With a first-hop redundancy protocol:

                Virtual Gateway
                192.168.10.1
                      |
             +--------+--------+
             |                 |
          Router 1          Router 2
           Active            Standby
             |                 |
             +--------+--------+
                      |
                     LAN
    

Hosts can use the virtual gateway rather than depending on the physical address of a single router.

What Happens During Failover?

1. The active device or path becomes unavailable.
2. The redundancy mechanism detects the failure.
3. An alternative device or path takes over.
4. Traffic is forwarded through the available path.
5. Network engineers can investigate the original failure.

   BEFORE FAILURE

   Users
     |
Virtual Gateway
     |
  Router A
   ACTIVE
     |
  Internet


   AFTER FAILURE

   Users
     |
Virtual Gateway
     |
  Router B
   ACTIVE
     |
  Internet
    

The exact convergence time depends on the technology, configuration, timers, network conditions, and failure type.

Other Sources of Redundancy

Practical Enterprise Example

                     INTERNET
                   /         \
                ISP 1       ISP 2
                  |           |
               Router 1    Router 2
                  \           /
                   \         /
                    Core
                   Switches
                  /         \
             Access 1      Access 2
                |             |
              Users         Users

          Monitoring + Alerts
                   |
              Network Team
    

This design has multiple layers of resilience:

• Multiple ISP connections
• Multiple routers
• Redundant switching paths
• Gateway redundancy
• Monitoring and alerting

The goal is not to guarantee that every failure is invisible. Instead, the architecture is designed so that individual failures have a reduced chance of causing a complete service outage.

Common Mistakes

• Adding redundant links without considering Layer 2 loops
• Assuming redundancy automatically means load balancing
• Not testing failover scenarios
• Failing to monitor redundant components
• Ignoring configuration consistency
• Having redundant devices but a single power source
• Creating redundancy without documenting the design

Quick Quiz

Question 1

What is failover?

Question 2

What is HSRP?

Question 3

What is VRRP?

Question 4

Why are redundant links useful?

Question 5

What is high availability?

Key Takeaways

• Failover moves traffic or service to an alternative when a failure occurs.
• Redundant links provide alternative network paths.
• High availability aims to maintain service availability despite failures.
• HSRP provides Cisco first-hop gateway redundancy.
• VRRP provides standards-based first-hop gateway redundancy.
• Virtual gateways allow hosts to use a resilient gateway address.
• STP/RSTP can help manage redundant Layer 2 paths.
• EtherChannel can combine multiple physical links into a logical link.
• Redundancy should be monitored and tested.