OCI Observability Professional study plan

Map every OCI Monitoring component (metrics, metric namespaces, dimensions, alarms, MQL) and write a one-page glossary in your own words

Enable the Monitoring service metrics for a compute instance and a load balancer, then locate them in the Metrics Explorer by namespace and dimension

Draft 15 MQL queries covering aggregation, groupBy, rate, and interval functions, and verify each returns data in the console

Configure a basic threshold alarm on CPU utilization and record the exact evaluation cadence and notification path it uses

Distinguish agent-based metrics (Compute Agent, Management Agent) from native service metrics and document which namespaces require which agent

Install and configure the Management Agent on a compute instance to emit custom metrics from a script

Build a custom metric namespace and publish at least three data points via the CLI or SDK

Create a composite alarm using multiple metrics and explain in writing how alarm state transitions behave

Explain the OCI Logging model (log groups, log types, service logs, custom logs, agent configurations) and sketch how they relate

Enable a VCN flow log and a service log for an Object Storage bucket, then query both in the console

Configure the unified monitoring agent to collect a custom application log file from a compute instance

Write five saved searches that filter by source, severity, and time window and note the search syntax rules

Connect OCI Logging to the Application Performance Monitoring (APM) domain by correlating trace IDs with log entries in a sample app

Explain the APM data model (domains, data keys, spans, traces, span logs) and write a short summary of each concept

Instrument a sample application with the APM Java or Python agent and confirm spans appear in the APM UI

Compare Logging Analytics with plain Logging and list three use cases where each is the better fit

Create an APM domain, generate a data key, and configure the agent with it in a fresh environment

Build a synthetic monitor against a public URL and configure a scripted browser monitor for a login flow

Interpret the APM dashboards: identify the slowest spans, error rates, and throughput and explain what each metric means

Configure a span filter or URL filter to reduce noise and verify the effect on trace volume

Diagram the Service Connector Hub model (source, task, target) and list valid source and target combinations from the docs

Create a connector that streams Object Storage logs to a Stream, then a second connector from the Stream to a Functions target

Explain how connectors handle failures, retries, and dead-letter behavior and document the settings that control them

Configure an alarm that fires into a Notification topic with an email subscription and confirm end-to-end delivery

Build a connector that routes Logging data to a Stream and then to a downstream analytics target, and measure latency

Design an alarm strategy covering availability, latency, and saturation, and implement at least four alarms across two services

Configure a Functions-based alarm handler that reacts to a notification and logs the event

Review IAM policies required for connectors, alarms, and topics and write the minimal policy statements needed

Re-take practice questions across all four domains and log every miss with the underlying concept

Re-derive the MQL and log search syntax from memory without consulting notes, then check against your cheat sheet

Explain aloud, in under five minutes each, how Monitoring, Logging, APM, and Service Connector Hub interoperate in one architecture

Close every gap flagged in weeks 1 through 7 by rebuilding the weakest lab from scratch