Google Professional Cloud DevOps Engineer Question 43
Select 3Google Cloud PlatformYou are a DevOps engineer managing a critical application running on Google Cloud. The application uses a managed instance group (MIG) with autoscaling enabled. You need to patch the VM instances with an important security update while ensuring zero downtime. Which of the following practices should you follow?
- A
Enable rolling updates for the MIG and configure the update policy to use 'proactive' replacement.
- B
Test the patch in a staging environment identical to production before applying it to the MIG.
- C
Manually stop all instances in the MIG, apply the patch, and restart them sequentially.
- D
Use canary updates by updating only a small subset of instances first and monitoring their behavior.
- E
Configure the MIG to use a 'recreate' update policy to ensure all instances are replaced at once.
Show answer and explanation
Correct answers: A, B, D
Explanation
To ensure safe and secure patching with zero downtime for a critical application, you should use rolling updates with a 'proactive' replacement strategy, test patches in a staging environment before production, and consider using canary updates to validate the patch on a small subset of instances. These practices minimize risk and maintain application availability while applying security updates.
- A. Correct.
Correct: Enabling rolling updates with 'proactive' replacement ensures that instances are updated in batches while maintaining availability. This minimizes the risk of downtime.
- B. Correct.
Correct: Testing the patch in a staging environment helps verify its stability and compatibility, reducing the risk of introducing issues into production.
- C. Incorrect.
Incorrect: Manually stopping all instances and patching them sequentially would lead to downtime, which is not acceptable for a critical application.
- D. Correct.
Correct: Using canary updates ensures that only a small subset of instances is updated first, allowing you to detect and mitigate any issues before rolling out the patch to the entire group.
- E. Incorrect.
Incorrect: Configuring the MIG to use a 'recreate' update policy would replace all instances at once, causing downtime for the application.