Google Professional Cloud DevOps Engineer Question 150
Select 3Google Cloud PlatformYour team has deployed a highly available web application on Google Kubernetes Engine (GKE). Recently, users have reported intermittent HTTP 500 errors. Upon investigation, you notice that the issue occurs when a specific service in your application is scaled up. You suspect it might be related to resource limits or readiness probes. What steps should you take to troubleshoot and resolve the issue?
- A
Check the Kubernetes pod logs for error messages during the scaling process.
- B
Verify if the readiness probe configuration is correctly defined and matches the application's behavior.
- C
Increase the CPU and memory resource limits for the affected service without analyzing the resource usage.
- D
Use GKE's monitoring and logging tools, such as Cloud Operations Suite, to identify resource bottlenecks or performance issues.
- E
Restart all the pods in the GKE cluster to resolve the issue.
Show answer and explanation
Correct answers: A, B, D
Explanation
To troubleshoot HTTP 500 errors in a GKE application, you should start by inspecting logs and verifying readiness probe configurations, as these are common sources of issues during scaling. Additionally, leveraging GKE's monitoring tools can help you identify resource constraints or performance bottlenecks. Arbitrary actions like increasing resource limits or restarting pods without understanding the root cause are not effective troubleshooting practices.
- A. Correct.
Checking the pod logs can provide insights into why the application is failing, such as errors during initialization or runtime issues.
- B. Correct.
Readiness probes ensure the application is ready to handle traffic. Misconfigured probes can cause HTTP 500 errors during scaling.
- C. Incorrect.
Increasing resource limits without analyzing the application's resource usage is not a recommended practice, as it may not address the root cause of the problem and could lead to resource wastage.
- D. Correct.
GKE's monitoring and logging tools allow you to analyze resource usage and detect potential bottlenecks, making them valuable for troubleshooting such issues.
- E. Incorrect.
Restarting all pods is a blunt approach and does not identify or address the root cause of the issue, making it an ineffective troubleshooting step.