🔁Failover

NEW

Automatic switching of service to a redundant system or site when the primary becomes unavailable.

DESCRIPTION

Failover mechanisms detect failures and redirect traffic or promote standby instances to restore service. Implementations range from DNS-based routing to load balancer health checks and database leader election. Active-passive designs promote a standby; active-active designs reroute around failures. Correctness depends on health probing, quorum, and preventing split-brain. Regular testing and automation are essential to meet recovery objectives.

ALTERNATIVE DEFINITIONS

Wikipedia

In computing, failover is switching to a redundant or standby computer server, system, or network upon the failure of, or abnormal termination of, the previously active application, server, system, or network.

Classic general definition emphasizing redundancy.

Source

AWS

Amazon Route 53 DNS failover automatically routes your users to an alternate location to avoid site outages.

Vendor-specific DNS failover automation.

Source

RELATED TERMS

High Availability Active-Passive Health Check Load Balancer Disaster Recovery Recovery Time Objective

COMPETING TERMS

Active-Active

KNOWLEDGE AREAS

Infrastructure Cloud Data Security

TARGET AUDIENCE

Cloud Architect DevOps Engineer IT Manager Solution Architect Chief Information Security Officer

EXTERNAL SOURCES

• Wikipedia - Failover
• Microsoft Learn - Azure resiliency and failover patterns
• Google Cloud - Design for availability and failover
• AWS - DNS failover configuration guide

TOP QUESTIONS

What triggers a Failover?

Health checks, timeouts, or quorum loss can trigger automatic switchover to standby or alternate paths.

Category: Example

How does Failover relate to High Availability?

Failover is a key mechanism within HA architectures to maintain service continuity during component failures.

Category: Example

What is a common mistake in Failover design?

Insufficient health checks or overly aggressive thresholds cause flapping and cascading failures.

Category: Mistakes

Is DNS-based Failover sufficient?

It’s simple but subject to DNS caching and TTL delays; combining with load balancers improves responsiveness.

Category: Opinion

What is Failback after a Failover?

It’s the controlled return to the primary once it’s healthy, requiring careful state reconciliation.

Category: Management

How do databases implement Failover?

Leader election among replicas, synchronous replication, and promotion mechanisms enable rapid role changes.

Category: Example

What is the risk of split-brain during Failover?

Two primaries can cause data divergence and corruption. Quorum and fencing mitigate this risk.

Category: Risks

How to test Failover regularly?

Run game days or chaos tests to verify detection, promotion, and client reconnection behavior.

Category: Management

What metrics define good Failover?

RTO, error rate during transition, and data consistency post-failover are key.

Category: Requirements