Avoid Single Points of Failure

A single point of failure is one component that can take down an entire service. Multiple servers do not create resilience if they still depend on the same switch, gateway, power supply, or fault domain.

A simple example

Two servers may provide compute redundancy, but if both connect to the same switch, that switch remains a single point of failure. If it fails, both servers lose connectivity.

Common single points of failure include:

• A single network switch
• A single gateway or uplink
• A single power supply or power source
• Multiple redundant systems placed in the same fault domain

Build real redundancy

Critical services should use multiple servers, network paths, and power sources wherever possible. Make sure the redundant components do not share the same underlying dependency.