Monitoring, Automation, and Cost Management
Monitoring, Automation, and Cost Management
This domain tests how you observe resources, automate operations, and control spend.
Core rules
- Monitoring collects metrics per resource and fires alarms when a metric crosses a threshold for a sustained period. Alarms publish to a notification topic; users must subscribe (email, SMS, or webhook) to receive alerts. Logging stores logs in log groups; Service Connector Hub moves logs or events to targets such as Object Storage or Functions.
- Automation options include OCI CLI and SDKs for scripts, Resource Manager for Terraform stacks, Functions for event-driven code, and DevOps for pipelines. The Console shows work requests for long-running actions.
- Cost Management: budgets set a monthly amount and alert rules (actual or forecast) at percentages. Budgets alert but do not stop resources. Cost analysis groups by compartment, tag, or service. Cost-tracking tags must be defined tags enabled for cost tracking.
- Tagging strategy: use defined tags for cost allocation and freeform tags for ad hoc notes. Tag defaults apply at compartment level.
Worked example
A team wants an alert when monthly spend is likely to exceed $5,000. Create a budget scoped to the tenancy or compartment for $5,000, add a forecast alert rule at 80% and an actual alert at 100%, then subscribe the FinOps email to the notification topic. The budget will notify but will not shut down instances, so add a separate automation such as a Function or scheduled CLI script if hard stops are required.
Traps
- Creating an alarm without a subscribed notification topic; no one receives the email.
- Expecting budgets to stop resources; they only alert.
- Using freeform tags for cost tracking; only defined tags enabled for cost tracking appear in cost reports.
- Confusing metrics (numeric time series) with logs (text events).
- Forgetting that alarm evaluation needs a sustained breach, not a single data point.