Skip to content

Health-Based Failover #519

Description

@shaneutt

When all endpoints in a cluster fail health checks, Praxis returns errors to clients. There is no way to route to a fallback backend or serve a static error page during complete outages.

Goal

Configurable backup pools that activate automatically when the primary cluster has no healthy endpoints.

Requirements

  • Per-cluster fallback_cluster configuration referencing another cluster
  • Automatic activation when all primary endpoints are unhealthy
  • Automatic deactivation when any primary endpoint recovers
  • Access log indication when fallback is active
  • Works for both HTTP and TCP listeners

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Type

    Projects

    Milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions