Course contentsShow
← Software Engineering
Lesson 1116 of 1,93333. Error Handling and ResilienceFree lesson

Fallacy 5: Topology Doesn't Change

Network topology evolves constantly; services must handle dynamic discovery, routing changes, and node failures.

Fallacy 5: Topology Doesn't Change

What You'll Learn

You'll understand why assuming network topology is static leads to brittle systems, and how to build services that adapt to evolving infrastructure.

The Fallacy Explained

Many developers design systems as if the network layout is fixed: "Service A talks to Service B at 192.168.1.50, and that's that." In reality, network topology is constantly changing. Servers get added or removed for scaling, VMs migrate between hosts, containers restart with new IP addresses, data centers go offline, load balancers reroute traffic, and DNS records update.

Hardcoding assumptions about where services live or how traffic flows creates fragility. When topology changes—and it will—your system breaks or degrades.

Building for Dynamic Topology

Service Discovery: Use dynamic service registries (like Consul, etcd, or Kubernetes DNS) rather than hardcoded addresses. Services register themselves on startup and deregister on shutdown.

Health-Aware Routing: Don't just know where services are—know which instances are healthy. Route traffic only to instances passing health checks.

Graceful Node Failures: Design for nodes disappearing without warning. Use retries with backoff, circuit breakers, and failover to healthy instances.

Configuration Updates: Support runtime configuration changes. When DNS changes or new nodes appear, your service should adapt without restarts.

Real-World Analogy

Think of a city where businesses constantly move locations. If your GPS app hardcoded every address, it would be useless within weeks. Instead, it queries a live database of locations. Your services need the same dynamic awareness.

Key Implications

  • Never hardcode IP addresses or hostnames
  • Implement service discovery and dynamic DNS resolution
  • Monitor topology changes and react automatically
  • Test how your system behaves when services move or disappear

Key Takeaway: Network topology is fluid, not frozen—design services with discovery mechanisms, health awareness, and automatic failover to handle constant infrastructure change.