Kubernetes_Liveness_Readiness_Probes

What it does

This skill allows AI agents to configure and understand Kubernetes liveness and readiness probes. It enables the agent to define how Kubernetes determines if a container is healthy (liveness) or ready to serve traffic (readiness), ensuring application stability and efficient resource utilization within a cluster. The skill facilitates automated probe configuration based on application requirements.

When to use it

• Automated Deployment: Configure probes during deployment of new applications to ensure they are properly monitored for health and readiness.
• Troubleshooting Application Issues: Analyze existing probe configurations to diagnose why an application might be unhealthy or unavailable.
• Optimizing Resource Allocation: Adjust probe thresholds to fine-tune when Kubernetes restarts failing containers or removes them from service rotation.
• Self-Healing Clusters: Implement proactive health checks that automatically restart or replace failing pods based on probe results.

Key capabilities

• Define liveness probes (HTTP GET, TCP socket, Exec command)
• Define readiness probes (HTTP GET, TCP socket, Exec command)
• Configure initial delay and period seconds for probes
• Understand probe failure thresholds

Example prompts

• "Create a readiness probe for my application that checks the /healthz endpoint."
• "What is the current liveness probe configuration for pod 'my-app'?"
• "Set the initial delay to 30 seconds and period to 10 seconds for the liveness probe of container 'web'."

Tips & gotchas

Ensure the probes accurately reflect the application’s health. Incorrectly configured probes can lead to unnecessary restarts or traffic disruption. Understanding your application's dependencies is crucial for defining effective readiness probes.