Examkin

Question

Difficulty: HardRelease Management and Deployment Strategies

A global media streaming platform hosts its microservices on Google Kubernetes Engine (GKE). The platform team is designing a progressive delivery strategy using Google Cloud Deploy to automate release rollouts across production clusters. The operational objective requires deploying new service revisions using a canary release pattern, gradually advancing traffic from 10% to 100%, while continuously validating application metrics against Cloud Monitoring. If the canary revision's HTTP 5xx error rate exceeds 1% during any evaluation phase, the deployment pipeline must automatically halt progress and roll back traffic to the previous stable release without manual intervention. Which architecture and release configuration best fulfills these requirements?

  1. Define a Cloud Deploy delivery pipeline with progressive canary targets and integrated Cloud Deploy verify jobs that execute automated metric checks against Cloud Monitoring, automatically invoking a pipeline rollback phase if pre-configured error thresholds are breached.Answer
  2. B
    Re-architect the streaming workload from GKE to App Engine Flexible Environment with traffic splitting enabled, relying on App Engine's built-in feature to automatically monitor container CPU utilization and execute rollbacks if metrics exceed safe limits.
  3. C
    Configure the External Application Load Balancer health check endpoints to perform deep SQL queries against the backing database, ensuring load balancer health checks mark canary pods as unhealthy whenever HTTP error rates increase.
  4. D
    Store the CI/CD pipeline deployment state files in local storage on the build server and run manual gcloud deployment commands sequentially across cluster regions to split traffic and evaluate error logs.

Answer

The optimal solution is to define a Cloud Deploy delivery pipeline with progressive canary targets and integrated Cloud Deploy verify jobs that query Cloud Monitoring metrics, automatically triggering pipeline rollbacks upon error threshold violations.
Google Cloud Deploy provides native support for canary release strategies and progressive delivery pipelines. By configuring Cloud Deploy verification steps (using Cloud Deploy verify), the pipeline can evaluate Cloud Monitoring metrics after each phase. If metrics such as HTTP 5xx error rates exceed designated thresholds, Cloud Deploy automatically halts promotion and rolls back traffic to the previous stable release, fully satisfying operational reliability requirements without manual intervention.

Step-by-Step Solution

1
Establish Progressive Delivery Targets
Cloud Deploy configures target phases (e.g., 10%, 25%, 50%, 100%) to route traffic safely to new application revisions.
Canary deployment strategies minimize blast radius by exposing a fraction of production traffic to new code releases.
2
Integrate Verification Jobs with Operational Monitoring
Cloud Deploy verify executes automated test suites or metric assertions querying Cloud Monitoring during each rollout phase.
Automated verification ensures empirical validation of service health (such as HTTP 5xx error rates) prior to promoting the release.
3
Configure Automated Rollback Rules
Upon detecting metric threshold breaches (e.g., error rate > 1%), Cloud Deploy halts advancement and reverts traffic to the last known good revision.
Automated rollback prevents sustained user impact and fulfills strict availability SLOs.

Key Concept

Progressive Delivery and Automated Verification with Google Cloud Deploy
Rate this question

Topics

Release Management and Deployment Strategies

More questions from this topic