Why should you care?

Observability stack lesson — Prometheus, Grafana, OTel, alert runbooks enterprise-wide.

See it live — copy this example

Use AWS/Azure/GCP free tier or local Docker/Kind. Sketches and YAML are meant to be typed and adapted.

# CloudVerse shared observability
# OTel collectors (DaemonSet) → Mimir + Loki + Tempo
# Grafana folders per squad; SSO via Entra ID
# Alertmanager routes: page vs ticket
# SLO recording rules exported to exec dashboard
groups:
- name: cloudverse-platform-slo
  rules:
  - record: slo:api_availability:ratio
    expr: 1 - (sum(rate(http_5xx[30d])) / sum(rate(http_total[30d])))

What happened?

• Platform team operates the stack; product teams own dashboards and runbooks for their services.
• Follow the steps below — typing the code yourself is the fastest way to learn.

Practice next

1. Deploy OTel collector chart.
2. Onboard one squad ServiceMonitor.
3. Create SLO dashboard.
4. Add log-based metric for security events.
5. Offer self-service dashboard templates.

Remember

Shared metrics/logs/traces. SLO rules centralized. Runbooks per service.

CloudVerse observability hub

Three squads need unified incident view.

Outcome: One Grafana org; MTTR drops 25%.