Architecting on AWS Certification Non-Destructive Testing Methods 5 — Questions and Answers
Question 1: A team wants to verify that CloudWatch alarms correctly trigger before deploying new auto-scaling policies to production. Which approach achieves this non-destructively?
- Manually scale the fleet down to zero and observe alarms
- Use the CloudWatch set-alarm-state API to force alarm states and verify downstream actions (Correct answer)
- Delete the existing alarms and recreate them with new thresholds
- Disable the auto-scaling group and test alarm math manually
Correct answer: Use the CloudWatch set-alarm-state API to force alarm states and verify downstream actions
set-alarm-state forces an alarm into OK, ALARM, or INSUFFICIENT_DATA state so you can verify that connected actions (SNS, Auto Scaling) fire correctly without real metric conditions.
Question 2: Which AWS X-Ray feature helps teams understand the impact of code changes on latency without introducing load into production?
- X-Ray groups and filter expressions applied to existing trace data from a staging environment (Correct answer)
- X-Ray sampling rules set to 100% on the production service
- X-Ray daemon log injection
- X-Ray encryption configuration audit
Correct answer: X-Ray groups and filter expressions applied to existing trace data from a staging environment
Analyzing X-Ray traces from a staging environment with filter expressions gives accurate latency profiling without any production risk.
Question 3: A DevOps team needs to validate that a new CodePipeline configuration correctly deploys to dev before it can be promoted. Which CodePipeline feature enforces this gate non-destructively?
- Pipeline version rollback
- Manual approval action between the dev deploy stage and the production deploy stage (Correct answer)
- CodePipeline parallel actions
- Artifact encryption with KMS
Correct answer: Manual approval action between the dev deploy stage and the production deploy stage
A manual approval action pauses the pipeline after dev deployment, allowing the team to validate the environment before the pipeline proceeds to production.
Question 4: An architect must confirm that Route 53 health checks correctly mark an endpoint unhealthy before switching DNS. What is the safest test method?
- Remove the DNS record and observe recovery time
- Configure a health check against a test endpoint that returns a non-200 status and verify the health check transitions to unhealthy (Correct answer)
- Disable Route 53 DNSSEC and monitor propagation
- Point all weighted routing records to zero weight
Correct answer: Configure a health check against a test endpoint that returns a non-200 status and verify the health check transitions to unhealthy
Testing health check logic against a dedicated test endpoint that deliberately returns an unhealthy response validates the behavior without touching live DNS records.
Question 5: Which AWS Config feature allows teams to evaluate whether resources comply with rules before those resources are provisioned, acting as a pre-deployment gate?
- AWS Config conformance packs
- AWS Config proactive evaluation mode (proactive rules) (Correct answer)
- AWS Config remediation actions
- AWS Config aggregator
Correct answer: AWS Config proactive evaluation mode (proactive rules)
Proactive evaluation mode lets you run Config rules against a CloudFormation template before deployment so non-compliant resources are caught before they are created.
Question 6: A team uses EventBridge to trigger downstream workflows. How can they test a new event pattern without publishing real events from the source application?
- Temporarily modify the source application to emit test payloads
- Use the EventBridge TestEventPattern API or the console's event sandbox to validate rules against sample JSON (Correct answer)
- Delete the rule and recreate it, then wait for the next real event
- Set the rule's event bus to a dead-letter queue and inspect messages
Correct answer: Use the EventBridge TestEventPattern API or the console's event sandbox to validate rules against sample JSON
The EventBridge event sandbox (and TestEventPattern API) lets you paste a sample event JSON and immediately see which rules would match, without publishing any real events.
Question 7: Which strategy allows a data engineering team to validate that a new Glue ETL job produces correct output before replacing the production job?
- Run the new job against production data and overwrite the output bucket
- Run the new Glue job in a separate workflow pointing to a copy of the source data and a separate output location, then compare results (Correct answer)
- Modify the existing production Glue job script in-place and run it
- Delete the existing Glue job and deploy the new one, relying on S3 versioning for rollback
Correct answer: Run the new Glue job in a separate workflow pointing to a copy of the source data and a separate output location, then compare results
Running the new job against a data copy in an isolated output location lets you compare results with the production job's output before making any cutover.
Architecting on AWS Certification Non-Destructive Testing Methods 5
A team wants to verify that CloudWatch alarms correctly trigger before deploying new auto-scaling policies to production.
Which approach achieves this non-destructively?