Google Professional Cloud DevOps Engineer Question 75
Select 3Google Cloud PlatformYour team recently deployed a new version of a containerized application to Google Kubernetes Engine (GKE). After the deployment, users report intermittent HTTP 500 errors. Upon investigation, you find that the new version uses a new configuration file that was not included in the container image. What steps should you take to troubleshoot and mitigate the issue?
- A
Roll back the deployment to the previous version using the Kubernetes Deployment rollback feature.
- B
Inspect the application logs to verify the cause of the HTTP 500 errors.
- C
Update the container image to include the missing configuration file and redeploy the application.
- D
Increase the resource limits for the containers to ensure they have enough capacity to handle traffic.
- E
Add a readiness probe to the deployment to ensure only healthy pods serve traffic.
Show answer and explanation
Correct answers: A, B, C
Explanation
When troubleshooting deployment issues, it is important to first mitigate user impact by rolling back to a stable version. To identify the root cause, application logs should be analyzed. Once the issue is identified (in this case, a missing configuration file), the permanent fix should be implemented, updating the container image and redeploying the application.
- A. Correct.
Rolling back the deployment to the previous working version is a valid mitigation strategy to restore service while you troubleshoot the issue.
- B. Correct.
Inspecting application logs can help confirm that the missing configuration file is the root cause of the errors.
- C. Correct.
Updating the container image to include the missing configuration file and redeploying the application will resolve the issue permanently.
- D. Incorrect.
Increasing resource limits is unrelated to the reported issue and would not resolve the HTTP 500 errors caused by the missing configuration file.
- E. Incorrect.
Adding a readiness probe can help ensure that only healthy pods serve traffic, but it does not address the root cause of the HTTP 500 errors in this scenario.