Failover
Failover Routing for Traffic Distributor
Failover routing protects an application by keeping one fully functional environment active and a second environment ready as a standby backup.
All incoming requests are sent to the primary backend while it remains healthy. When the primary service becomes unavailable, Traffic Distributor redirects requests to the backup backend automatically.
On this page
How Failover Routing Works
- One backend is selected as the primary application environment.
- The second backend remains available as a standby copy.
- Traffic Distributor sends all requests to the primary backend.
- Health checks monitor whether the primary service remains available.
- When the primary backend fails, incoming requests are redirected to the backup.
The backup environment should remain fully operational and ready to serve production traffic whenever failover occurs.
Traffic Ratio
Failover does not support a custom weighted distribution between both backends. The routing ratio is always configured as 100% / 0%.
100%
Primary Backend
Receives all incoming requests while it remains available.
0%
Backup Backend
Receives no normal traffic and is used only when the primary service fails.
One active backend: At any moment, requests are handled by either the primary backend or the backup backend, not by both simultaneously.
Cross-Region Protection
Placing the primary and backup environments in different infrastructure regions can protect the application from failures tied to one hardware location.
- Primary and backup copies can run on separate hardware.
- A region-level outage does not necessarily affect both backends.
- The backup environment can continue serving users when the primary location is unavailable.
Application Availability
Customer requests are automatically redirected to the working backend. When the backup is healthy and properly synchronized, users may experience little or no visible interruption during failover.
Actual failover time depends on health-check frequency, timeout settings, DNS and connection behaviour, and how quickly the backup application becomes ready to serve requests.
Configuration Recommendations
- Keep the backup application, configuration, and dependencies ready for production use.
- Synchronize required files and databases between both environments.
- Use separate regions when protection from hardware or location failure is required.
- Test failover regularly by temporarily making the primary backend unavailable.
- Monitor health-check results and backend availability.
- Verify that the backup environment has enough resources for the complete production workload.
Data consistency: Failover routing redirects traffic but does not synchronize application data automatically. Use replication, shared storage, or another synchronization method for stateful workloads.