Google Professional Cloud DevOps Engineer Question 146
Single answerGoogle Cloud PlatformYour team manages a critical e-commerce application running on Google Cloud. To ensure the application meets its Service Level Objectives (SLOs), you want to configure alerting in Cloud Monitoring. The business has set the following requirements:
- Customers must experience less than 1% error rates over a rolling window of 30 minutes (SLO).
- The team must be alerted only if the error rate exceeds the threshold for at least 10 minutes.
How should you configure an alerting policy in Cloud Monitoring to meet these requirements?
- A
Create an alerting policy with a metric-based condition monitoring HTTP error rates. Set the condition to trigger if the error rate exceeds 1% for at least 10 minutes.
- B
Create an alerting policy with a log-based condition monitoring error messages in Cloud Logging. Set the condition to trigger if the number of error logs exceeds a fixed count within 10 minutes.
- C
Create an alerting policy with a metric-based condition monitoring HTTP latency. Set the condition to trigger if the latency exceeds 1 second for at least 10 minutes.
- D
Create an alerting policy with a metric-based condition monitoring CPU utilization. Set the condition to trigger if CPU usage goes above 80% for at least 10 minutes.
Show answer and explanation
Correct answer: A
Explanation
To meet the SLO of keeping error rates below 1%, it's essential to monitor HTTP error metrics with a metric-based condition in Cloud Monitoring. The condition must be configured to trigger alerts when the error rate exceeds 1% for more than 10 minutes, as specified in the requirements. This ensures timely and accurate alerts while avoiding false positives caused by short-term fluctuations in error rates.
- A. Correct.
This option correctly defines an alerting policy that monitors HTTP error rates and ensures alerts are triggered if the error rate exceeds 1% for at least 10 minutes, meeting the SLO requirements.
- B. Incorrect.
This option does not align with the requirements because it uses a log-based condition instead of monitoring metrics. Additionally, monitoring a fixed count of error logs does not directly ensure adherence to the error rate SLO.
- C. Incorrect.
This option is incorrect because it focuses on HTTP latency, which is unrelated to the error rate SLO specified in the requirements.
- D. Incorrect.
This option is incorrect because monitoring CPU utilization does not address the requirement of ensuring the error rate remains under 1%.