Kubernetes RCA & Alerting: Gemini, Loki, Prometheus, Slack

This n8n workflow streamlines Kubernetes incident response by integrating observability tools like Loki for logs, Prometheus for metrics, and Google Gemini for AI-driven root cause analysis (RCA). It monitors pod states, filters critical errors, enriches data with Kubernetes documentation, and sends deduplicated, formatted alerts to Slack channels. Ideal for production-grade clusters, it handles crashloops, restarts, and health checks automatically.

Key features include HTTP requests to Loki and Prometheus for data collection, Kubernetes SSH for pod status verification (ignoring terminated pods), and structured prompts to Gemini for actionable insights. Post-RCA enrichment adds relevant docs links, ensuring comprehensive troubleshooting. Benefits include slashed manual debugging time, faster MTTR, and plug-and-play setup for SREs and DevOps engineers.

Use cases: Automate alerting for pending/not-ready pods, node failures, or service issues in multi-namespace environments. Perfect for teams leveraging n8n with Slack, Loki, and Prometheus stacks, delivering business value through proactive monitoring and intelligent automation.