10.1 · Capability in Operations

Product observability

Seeing what the running system is doing, so problems are found by the team before users report them.

Map6 skills

Not startedLearningComfortable

6 skillsClick a hexagon to set your status

10.1.1LogsStructured records of what the system did.
10.1.2TracesFollowing a single request across services and components.
10.1.3Service metricsNumbers that describe system health, such as error rate and latency.
10.1.4DashboardsShared views of the metrics that matter.
10.1.5AlertingNotifying people when something needs attention.
10.1.6Performance monitoringTracking real-world speed and errors as users experience them.

Mark all 6

Where to start

3 resources

docsStart here
Observability primer ↗
OpenTelemetry · 10 min
Logs, metrics and traces explained, and how they connect so you can answer questions you did not plan for.
bookStart here
Monitoring distributed systems (Site Reliability Engineering, chapter 6) ↗
Google SRE · 20 min
The four golden signals and the case for alerting on symptoms users feel rather than on every cause.
bookGo deeper
Monitoring (The Site Reliability Workbook, chapter 4) ↗
Google SRE · 20 min
Practical monitoring setups, from choosing metrics to moving from an alert to its cause.

4 tools

ToolOpen source
OpenTelemetry
opentelemetry.io ↗
Open standard and SDKs for emitting traces, metrics and logs to any backend.
ToolOpen source
Grafana
grafana.com ↗
Open-source dashboards for metrics, logs and traces from almost any data source.
ToolOpen source
Prometheus
prometheus.io ↗
Open-source metrics database and alerting, the common default for service metrics.
ToolFreemium
Sentry
sentry.io ↗
Error tracking and performance monitoring that groups exceptions and links them to releases.