How to create effective Grafana dashboards?

Question

Accepted Answer

**Grafana Dashboard** — metrics visualization for monitoring.

**Structure:**
1. Overview: SLI, errors, latency.
2. Services: RPS, errors, latency per service.
3. Infrastructure: CPU, RAM, disk.
4. Business: conversion, revenue.

**Useful PromQL:**
```promql
# Request Rate
sum(rate(http_requests_total[5m])) by (service)

# Error Rate
100 * sum(rate(http_requests_total{status=~"5.."}[5m])) /
     sum(rate(http_requests_total[5m]))

# P95 Latency
histogram_quantile(0.95,
  sum(rate(http_request_duration_seconds_bucket[5m])) by (le, service)
)

# Memory Usage
container_memory_working_set_bytes{namespace="production"} /
container_spec_memory_limit_bytes{namespace="production"} * 100
```

**Best practices:** Single pane of glass, {{namespace}} variables, deploy annotations, threshold (green/yellow/red).

DevOps: How to create effective Grafana dashboards?