3.7. High Availability

💡 First Principle: High availability is about surviving failure, not preventing it. Hardware will fail; software will crash; networks will have outages. HA architecture ensures that when individual components fail, the service continues. The question is never "will this fail?" but "when it fails, what happens?"

This reframing is critical for exam scenarios. When a question asks about high availability design, you're looking for the answer that eliminates single points of failure and ensures automatic failover.

⚠️ Common Misconception: Active-passive clustering wastes the passive node's resources because it "does nothing." The passive node is actively monitoring the heartbeat, maintaining state synchronization, and remaining ready for instant failover—it's a dedicated standby, not an idle machine. Active-active maximizes utilization but introduces coordination complexity.