Course contentsShow
← System Design
Lesson 81 of 1,9193. Load BalancingFree lesson

Single Point of Failure: Load Balancer HA

Learn how load balancers themselves can become bottlenecks and strategies for high availability.

Single Point of Failure: Load Balancer HA

What you'll learn: How load balancers themselves can become single points of failure and the strategies to prevent this.

The Irony of Load Balancers

You've introduced a load balancer to distribute traffic across multiple servers, eliminating the single point of failure from having just one server. But now there's a new problem: what if the load balancer itself goes down? Your entire system becomes unreachable, even though all your backend servers are perfectly healthy.

Think of it like a busy restaurant with multiple chefs (servers) but only one host at the entrance (load balancer). If that host is absent, customers can't get seated even though the kitchen is ready to cook.

High Availability Strategies

Active-Passive (Primary-Secondary):

  • Deploy two load balancers: one active, one standby
  • The passive load balancer monitors the active one using health checks
  • If the active fails, the passive takes over (failover)
  • Simple but wastes the passive load balancer's capacity during normal operation

Active-Active:

  • Both load balancers actively handle traffic simultaneously
  • They share the load using DNS round-robin or another mechanism
  • If one fails, the other absorbs all traffic
  • Better resource utilization but more complex configuration

DNS-Level Distribution:

  • Use multiple load balancer IP addresses in DNS records
  • Clients automatically try alternatives if one is unreachable
  • Combines well with active-active setups

Health Checks and Monitoring:

  • Load balancers must monitor each other
  • Automated failover triggers within seconds, not minutes
  • External monitoring ensures the monitoring system itself hasn't failed

Key Takeaway: Load balancers solve availability problems but create their own single point of failure. High availability requires redundant load balancers in active-passive or active-active configurations, with health checks and automatic failover mechanisms to ensure the traffic distribution layer never becomes the bottleneck.