CloudWatch¶
Note: Full content coming. Placeholder based on apprenticeship notes.
Ultra-Short Summary¶
CloudWatch is AWS's native monitoring and observability service. It collects metrics from all AWS services automatically, lets you create custom metrics, and triggers alarms that can auto-scale, notify, or remediate.
Key Components¶
| Component | What It Does |
|---|---|
| Metrics | Numeric data over time (CPU, latency, errors) |
| Logs | Text log streams from EC2, Lambda, containers |
| Alarms | Trigger action when metric crosses a threshold |
| Dashboards | Visual display of metrics |
| Logs Insights | SQL-like query language for log data |
| Events / EventBridge | React to state changes in AWS services |
Common Exam Patterns¶
- EC2 doesn't send memory/disk metrics by default — needs CloudWatch Agent
- CloudWatch Logs can stream to Lambda, Kinesis, or OpenSearch
- Alarms can trigger: SNS, Auto Scaling, EC2 actions
- Metrics retention: high-res (1s) for 3 hours → standard (1m) for 15 days → 5m for 63 days → 1h for 15 months
Expand this note after ILT monitoring session.