CloudWatch, X-Ray, CloudTrail, Config, Systems Manager, CloudFormation, Trusted Advisor and AWS Health in thirty-seven slides: metrics, their dimensions, resolution and retention, alarms and their actions; Logs, Logs Insights, metric filters, the CloudWatch agent and dashboards; traces, and the move from the X-Ray SDKs to OpenTelemetry and Application Signals; CloudTrail and Config side by side with CloudWatch; Session Manager, Run Command, Patch Manager and Parameter Store; CloudFormation templates, stacks, intrinsic functions, change sets, drift, StackSets and deletion policies; Trusted Advisor and the Health Dashboard. Every section ends with five exam questions.
The slides
The tool is thirty-seven slides: an introduction, then seven sections, listed on its left. Every section ends with five
exam questions, each with its answer and the tempting wrong answer.
- 0. Introduction
- Observability and infrastructure as code: the project, and the services of the session around it.
- The three questions about a running system: is it healthy, who did it, what changed.
- 1. Metrics and alarms
- A CloudWatch metric: namespace, name, dimensions, statistics.
- Resolution, retention and detailed monitoring: how long each period is kept.
- A CloudWatch alarm: period, evaluation periods, datapoints to alarm, its three states.
- What an alarm can do: SNS, Auto Scaling, EC2 actions, composite alarms.
- 2. Logs, the agent and dashboards
- CloudWatch Logs: log groups, streams, retention, sources.
- CloudWatch Logs Insights: a query over nginx's access log.
- A metric filter: log lines become a metric and an alarm.
- The CloudWatch agent: memory, disk and log files from inside the instance.
- A CloudWatch dashboard: the project's widgets.
- 3. Tracing
- A trace of one request: spans through the load balancer, the web server and two databases.
- X-Ray, OpenTelemetry and Application Signals: AWS's current guidance on instrumenting.
- 4. CloudTrail and Config
- AWS CloudTrail: one event record, Event history and trails, data events.
- AWS Config: a security group's configuration items, and a rule's verdicts.
- CloudTrail, CloudWatch and Config side by side: one change as each sees it.
- 5. Systems Manager
- AWS Systems Manager: a managed node and its three needs.
- Session Manager, Run Command, Patch Manager and Parameter Store.
- 6. CloudFormation
- A CloudFormation template, a stack, intrinsic functions and outputs.
- A change set, drift detection and StackSets.
- Protecting a stack's resources: deletion policies, stack policies, termination protection, rollback triggers.
- 7. Trusted Advisor and Health
- AWS Trusted Advisor: its six categories, and the checks each support plan gets.
- AWS Health: service health and account health, and its events in EventBridge.
The buttons on the left and right edges move between slides. So do the arrow keys and a swipe.
The lab
CloudFormation and CloudWatch Lab: the Project as a Stack practices
it.
Terraform: Code, Cloud and State shows the other common tool for
infrastructure as code.
The retention periods, the 30 dimensions, the high-resolution periods, the Parameter Store tiers, the X-Ray SDKs'
maintenance date (2026-02-25) and the checks of each support plan were read in AWS's documentation on 2026-11-16. The
account ID, instance IDs and addresses are placeholders: AWS's 111122223333 and i-1234567890abcdef0, and the
documentation ranges. The dashboard's lines and the Trusted Advisor counts are pictures, not data.
Sources