Google ACE Practice Question: Ensuring Successful Operation of a Cloud Solution

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Uptime check alert

  Why it's wrong here

  An uptime check alert only probes whether an external endpoint responds (e.g., HTTP 200), which measures availability, not application-level error rates or internal failures. A site reliability engineer needing immediate notification about elevated error rates cannot use uptime checks because they do not evaluate aggregated metrics such as '5xx response ratio' or 'exception count'. This option is therefore incorrect for detecting a trending error-rate spike.

• ✗

  Pub/Sub notification hook

  Why it's wrong here

  Pub/Sub is a fully managed asynchronous messaging service used to decouple services, not a native Cloud Monitoring alerting mechanism. Although you could publish alert notifications to Pub/Sub as a delivery channel, Pub/Sub itself does not evaluate metrics or trigger alerts; it only transports messages once an alert has already fired. Therefore, selecting a 'Pub/Sub notification hook' as the method to be notified would not perform the necessary metric evaluation and is not a valid alerting policy type.

• ✓

  Metric threshold alert

  Why this is correct

  A metric threshold alert continuously evaluates a time-series metric (e.g., error rate, request latency, or CPU utilization) against a user-defined threshold, such as 'error rate > 5% for 5 minutes', and immediately triggers a notification when the condition is met. This is precisely the right tool for an SRE who needs to be notified when application error rates exceed an acceptable level, because it supports real-time aggregation, sliding windows, and alerting policies with multiple notification channels. It is the correct answer.

• ✗

  Log-based alert (log metric trigger)

  Why it's wrong here

  A log-based alert (log metric trigger) is designed to fire based on the presence or count of specific log entries, such as a particular error string appearing in logs, and uses a logging metric (e.g., counter-based) to trigger the alert. However, it does not natively compute a time-series ratio like 'error rate per total requests' over a sliding window; you would need to build a custom log metric with advanced filters and potentially use logging-based threshold conditions. For an immediate, aggregated metric-threshold notification, this approach is less direct and more limited than a true metric threshold alert.

Go deeper

Related to this question