قابلية المراقبة باستخدام التسجيل المنظم والتتبع

Observability rests on logs, metrics, and traces. Together they answer what happened, how much, and where the time went across your serverless system.

Plain text logs are hard to search. Structured (JSON) logs let CloudWatch Logs Insights filter and aggregate by field, turning logs into queryable data.

Emit JSON with consistent fields so you can query by request, level, and custom attributes.

Pass a correlation ID through every service and log it everywhere. This lets you trace a single user request across multiple functions and queues.

CloudWatch Logs Insights runs SQL-like queries over structured logs.

AWS X-Ray records a trace as a request flows through Lambda, API Gateway, DynamoDB, and more, showing a timeline of every segment.

Turn on active tracing for the function, then instrument the SDK so downstream calls become subsegments.

Wrap your own logic in a subsegment to measure how long a specific block takes inside the trace.

The Embedded Metric Format lets you emit metrics inside a structured log line. CloudWatch extracts them automatically — no extra API call.

Create CloudWatch alarms on error rate, p99 latency, and DLQ depth. Alarms turn passive metrics into proactive alerts before users notice.

Log with intent:

• Use levels (DEBUG/INFO/ERROR) and a log-level env var
• Never log secrets or full PII
• Sample high-volume traces to control cost

You learned production observability:

• Logs, metrics, and traces are the three pillars
• Use structured JSON logs and correlation IDs
• X-Ray gives distributed traces across services
• Emit custom metrics with EMF and alarm on them