Cloud Engineer Cloud Monitoring, Logging & Observability 2 — Questions and Answers
Question 1: Which tool in Google Cloud is best suited for identifying performance bottlenecks in distributed microservices by visualizing request latency across services?
- Cloud Monitoring
- Cloud Profiler
- Cloud Trace (Correct answer)
- Error Reporting
Correct answer: Cloud Trace
Cloud Trace collects latency data from applications and displays distributed traces, allowing you to identify performance bottlenecks across microservices.
Question 2: A developer wants to be automatically notified when their application starts throwing a new type of exception in production. Which service provides this capability?
- Cloud Monitoring
- Cloud Logging
- Error Reporting (Correct answer)
- Cloud Trace
Correct answer: Error Reporting
Error Reporting aggregates and de-duplicates application errors and sends notifications when new error groups are detected in your application.
Question 3: You need to create a metric based on the count of log entries containing the word 'ERROR' in your application logs. What should you configure in Cloud Logging?
- Log sink
- Log-based metric (Correct answer)
- Log bucket
- Log exclusion
Correct answer: Log-based metric
Log-based metrics extract numeric values from log entries matching a filter and make them available as Cloud Monitoring metrics for alerting and dashboards.
Question 4: Which IAM role grants a user read-only access to view logs in Cloud Logging without allowing them to modify log sinks or exclusions?
- roles/logging.admin
- roles/logging.logWriter
- roles/logging.viewer (Correct answer)
- roles/logging.configWriter
Correct answer: roles/logging.viewer
The roles/logging.viewer role grants permission to view logs, log-based metrics, and logging configuration but not to modify them.
Question 5: A Cloud Monitoring alerting policy is configured with a condition duration of 5 minutes. What does this mean?
- The alert fires immediately when the threshold is exceeded for any instant
- The alert fires only if the threshold is continuously exceeded for 5 minutes (Correct answer)
- Notifications are delayed by 5 minutes after the alert fires
- The alert policy is evaluated every 5 minutes
Correct answer: The alert fires only if the threshold is continuously exceeded for 5 minutes
The condition duration (also called the alignment period or 'for' window) requires the threshold to be continuously violated for the specified duration before the alerting policy fires.
Question 6: You want to reduce Cloud Logging ingestion costs by dropping noisy DEBUG-level log entries from a specific service before they are stored. What should you configure?
- Log sink to /dev/null
- Log exclusion filter (Correct answer)
- Log bucket with reduced retention
- Cloud Monitoring alert suppression
Correct answer: Log exclusion filter
Log exclusion filters in Cloud Logging drop matching log entries before they are ingested and stored, reducing ingestion volume and costs.
Question 7: Which Cloud Profiler feature helps reduce the performance overhead of continuous profiling in production?
- It only profiles code during scheduled maintenance windows
- It uses statistical sampling to collect performance data with low overhead (Correct answer)
- It requires a debug build of the application
- It profiles only the main thread of the application
Correct answer: It uses statistical sampling to collect performance data with low overhead
Cloud Profiler uses statistical (sampling-based) profiling, which introduces minimal CPU and memory overhead, making it safe to run continuously in production.
Cloud Engineer Cloud Monitoring, Logging & Observability 2
Which tool in Google Cloud is best suited for identifying performance bottlenecks in distributed microservices by visualizing request latency across services?