Monitor K8s Services & Pods with Prometheus Alerts to Slack

This n8n workflow automates Kubernetes monitoring by querying Prometheus metrics every 5 minutes for critical issues: pods stuck in Pending state (due to scheduling problems like taints or capacity), containers failing readiness probes (Not Ready), sudden restart spikes over a 5-minute window, and service endpoint discovery changes (count diffs from snapshots). It aggregates findings into clean, emoji-enhanced Slack messages with full context on namespaces, pods, and services, including a summary block to minimize alert fatigue.

Benefits include real-time SRE/DevOps observability without custom scripts, reducing MTTR for cluster issues. It leverages kube-state-metrics for reliable pod/service data, supports optional Prometheus auth, and requires only Slack credentials. Ideal for production K8s environments to catch problems early.

Use cases: Multi-tenant clusters needing proactive alerting, hybrid cloud setups with Prometheus federation, or teams shifting left on ops with automated notifications. Outputs reduce noise while providing actionable details for quick triage.