2.2 Implementing and Managing Pipelines PROFESSIONAL-CLOUD-DEVOPS-ENGINEER Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   Which practice is enabled by feature flags to reduce the risk of releasing a new feature?

   • ADatabase schema migration
   • BGradual rollout to a subset of users
   • CAutomatic infrastructure scaling
   • DImmediate deletion of old code
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A security audit requires you to provide a detailed report of all deployment actions performed on a specific Artifact Registry repository in the last 30 days. The report must include who initiated the deployment, the source of the artifact, and the target environment. You have Cloud Audit Logs enabled. What is the most efficient way to retrieve this information?

   • AUse the Cloud Deploy console to view the delivery pipeline history and filter by the target environment.
   • BUse the Artifact Registry API to list all versions of the repository and check the 'createTime' field.
   • CUse the gcloud logging read command to query Cloud Audit Logs for the 'artifactregistry.googleapis.com' service and filter by the repository name.
   • DUse the Logs Explorer to query Cloud Audit Logs for 'deployment' events across 'cloudbuild.googleapis.com', 'clouddeploy.googleapis.com', and 'artifactregistry.googleapis.com' services, filtering by the repository name and time range.
   Reveal Solution Discussion Select an answer first
4. 8foundation · easy

   What is the primary purpose of a feature flag in progressive delivery?

   • ATo encrypt application data
   • BTo replace the need for version control
   • CTo automatically rollback failed deployments
   • DTo control feature exposure independently of deployment
   Reveal Solution Discussion Select an answer first
5. 9application · medium

   You have deployed a new version of a machine learning model to production. The model's offline accuracy is high, but you are concerned about data drift. You want to monitor the model's performance in production. Which approach is the most appropriate for detecting data drift?

   • ACompare the distribution of the model's input features in production to the distribution of the training data.
   • BMonitor the model's prediction accuracy on a sample of production data that is labeled by a human reviewer.
   • CMonitor the model's prediction latency in production.
   • DMonitor the model's CPU and memory utilization.
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   If a canary deployment shows increased latency for a small subset of users, what is an appropriate immediate mitigation?

   • AIgnore the latency and continue rollout
   • BShift all traffic back to the stable version
   • CDelete the canary version
   • DIncrease canary traffic to test more
   Reveal Solution Discussion Select an answer first