gke-reliability

What problem does it solve?

GKE workloads frequently experience avoidable downtime from unplanned voluntary disruptions, missing health checks, and poor failure domain distribution, leading to service outages and degraded user experience for production applications.

Core Features & Use Cases

• Pod Disruption Budgets (PDBs): Configure minimum availability guarantees for workloads during voluntary disruptions like node upgrades and autoscaler scale-down events.
• Health Probe Setup: Implement liveness, readiness, and startup probes to validate container health, manage traffic routing, and prevent premature restarts for slow-starting applications.
• Failure Domain Resilience: Use topology spread constraints and replica best practices to distribute workloads across zones and nodes, surviving zone and node failures.
• Use Case: A team running production stateless APIs on GKE can use this skill to implement all required reliability configurations to maintain 99.9% uptime during cluster maintenance and zone outages.

Quick Start

Use the gke-reliability skill to configure production-grade reliability settings for your GKE deployment, including PDBs, health probes, and zone spread constraints to eliminate avoidable downtime.