Service Transparency

Metrics

  • Host Level
  • Infrastructure Level
  • Service Level

Host Level

  • Cpu
  • Memory
  • Disk
  • ...

Infrastructure Metrics

  • Nginx
  • MessageQueue
  • postgresql?
  • ....

Service Metrics

  • Uptime
  • Running Jobs
  • Response Time
  • Number of Errors
  • ...

Product Metrics

  • Download
  • Purchases
  • Views
  • ...

Alerting

Metrics Sudden Changes

Emergency ABC

False Alerts

Dashboard

  • Minimum sufficient data
  • Metrics with alerts
  • Pipeline

Log

  • What to Log?
  • What not to Log?
  • Quick Search on
    • Time
    • Event

Kibana

Error

  • Resolve Errors!
  • Severity
  • Sentry

Tracing

  • Network Latency
  • Bottlenecks
  • Zipkin/Jaeger

OpenTracing

Made with Slides.com