1Z0-1104-25 Question 166
Single answerYou manage a fleet of Compute instances running critical workloads in Oracle Cloud Infrastructure (OCI). Your security policy requires that you monitor and receive alerts when any instance� memory usage exceeds 80% for at least 10 consecutive minutes. Which approach should you take to meet this requirement using OCI Monitoring?
- A
Enable the default OCI Monitoring metrics for CPU utilization and create an alarm for 80% CPU usage over 10 minutes.
- B
Use the Oracle Cloud Agent to collect memory usage, publish these metrics to OCI Monitoring, and configure an alarm with an 80% threshold for 10 consecutive minutes.
- C
Rely on Oracle Cloud Infrastructure Logging to capture instance console output and manually parse logs for memory usage spikes.
- D
Export detailed OS-level statistics to an external analytics tool, then manually generate alerts based on memory thresholds.
Show answer and explanation
Correct answer: B
Explanation
OCI supports out-of-the-box metrics for CPU and network usage on Compute instances. Memory usage, however, requires installing and configuring the Oracle Cloud Agent to publish custom metrics to OCI Monitoring. Once memory metrics are available, you can configure an alarm to automatically notify you when memory usage exceeds the specified threshold for the defined interval. Refer to the OCI documentation on 'Monitoring Compute Instances' for detailed instructions on configuring these custom metrics and alarms.
- A. Incorrect.
Incorrect. CPU metrics are available by default, but they do not capture memory usage. Monitoring memory usage specifically requires custom or agent-based metrics.
- B. Correct.
Correct. Memory usage is not a default OCI metric. By installing and configuring the Oracle Cloud Agent, you can collect and publish memory usage data as custom metrics, then set an alarm for when it exceeds 80% for 10 minutes.
- C. Incorrect.
Incorrect. Logging can capture syslog or console output, but it is not designed for real-time numeric threshold analysis of memory usage. You would still need to process logs separately for thresholds.
- D. Incorrect.
Incorrect. While external analytics tools can process OS-level data, this approach bypasses OCI Monitoring� built-in alarm capabilities, delaying real-time alerts and complicating your security posture.