Real-time Kubernetes CPU Spike Alerts: Prometheus to Slack

This workflow automates real-time monitoring of CPU usage in Kubernetes pods across EKS, GKE, or AKS clusters using Prometheus queries. It polls metrics like container_cpu_usage_seconds_total and kube_pod_container_resource_limits every 5 minutes, calculates usage rates, and flags any pod exceeding a configurable threshold (e.g., 0.8 cores). Pods are intelligently grouped by application name to consolidate alerts and minimize notification fatigue, providing structured details on affected apps, pods, and usage levels.

Ideal for DevOps, SRE, and platform teams, it delivers plug-and-play observability without needing Alertmanager or complex setups. Benefits include instant visibility into resource spikes for apps like Argo CD, Loki, or Promtail, enabling quick remediation. It's fully no-code, extensible to memory, disk, or network metrics, and saves hours of manual scripting or dashboard watching.

Setup is straightforward: configure your Prometheus URL, add a Slack bot token (chat:write scope), import the workflow, and tweak threshold, channel, or cron schedule. Requires a Prometheus instance with kube-state-metrics and an n8n setup (self-hosted or cloud). Customize queries for multi-cluster support or integrate with PagerDuty for escalation.

Tags: Prometheus, Slack, Kubernetes, Alert, n8n, DevOps, Observability, CPU Spike, Monitoring