Exam AWS DevOps Engineer Professional All Questions

View all questions & answers for the AWS DevOps Engineer Professional exam

Exam AWS DevOps Engineer Professional topic 1 question 51 discussion

Exam question from Amazon's AWS DevOps Engineer Professional

Question #: 51
Topic #: 1

[All AWS DevOps Engineer Professional Questions]

An online company uses Amazon EC2 Auto Scaling extensively to provide an excellent customer experience while minimizing the number of running EC2 instances. The company's self-hosted Puppet environment in the application layer manages the configuration of the instances. The IT manager wants the lowest licensing costs and wants to ensure that whenever the EC2 Auto Scaling group scales down, removed EC2 instances are deregistered from the Puppet master as soon as possible.
How can the requirement be met?

A. At instance launch time, use EC2 user data to deploy the AWS CodeDeploy agent. Use CodeDeploy to install the Puppet agent. When the Auto Scaling group scales out, run a script to register the newly deployed instances to the Puppet master. When the Auto Scaling group scales in, use the EC2 Auto Scaling EC2_INSTANCE_TERMINATING lifecycle hook to trigger de-registration from the Puppet master.
B. Bake the AWS CodeDeploy agent into the base AMI. When the Auto Scaling group scales out, use CodeDeploy to install the Puppet agent, and execute a script to register the newly deployed instances to the Puppet master. When the Auto Scaling group scales in, use the CodeDeploy ApplicationStop lifecycle hook to run a script to de-register the instance from the Puppet master.
C. At instance launch time, use EC2 user data to deploy the AWS CodeDeploy agent. When the Auto Scaling group scales out, use CodeDeploy to install the Puppet agent, and run a script to register the newly deployed instances to the Puppet master. When the Auto Scaling group scales in, use the EC2 user data instance stop script to run a script to de-register the instance from the Puppet master.
D. Bake the AWS Systems Manager agent into the base AMI. When the Auto Scaling group scales out, use the AWS Systems Manager to install the Puppet agent, and run a script to register the newly deployed instances to the Puppet master. When the Auto Scaling group scales in, use the Systems Manager instance stop lifecycle hook to run a script to de-register the instance from the Puppet master.

Show Suggested Answer

Suggested Answer: A 🗳️

by marwan at Nov. 13, 2019, 1:29 p.m.

Disclaimers:

- ExamTopics website is not related to, affiliated with, endorsed or authorized by Amazon.
- Trademarks, certification & product names are used for reference only and belong to Amazon.

Comments

Submit Cancel

pleasespammelater

Highly Voted 3 years, 7 months ago

This is a tough one. The key to understanding it is that CodeDeploy and ASG's can work together! There's a great blog post about this: https://aws.amazon.com/blogs/devops/under-the-hood-aws-codedeploy-and-auto-scaling-integration/ Here is my analysis: A: This is the correct approach and matches the blog post writeup B: The CodeDeploy ApplicationStop lifecycle hook relates to upgrading an instance in place. The EC2 lifecycle hook is needed here. C: There is no such thing as a "user data instance stop script" D: SSM does not have lifecycle hooks

upvoted 53 times

...

amzngenius

Highly Voted 3 years, 8 months ago

D： https://aws.amazon.com/cn/blogs/mt/configure-amazon-ec2-instances-in-an-auto-scaling-group-using-state-manager/

upvoted 12 times

...

m4r0ck

Most Recent 1 year, 10 months ago

Selected Answer: A

A is the correct answer. ref. https://docs.aws.amazon.com/autoscaling/ec2/APIReference/API_LifecycleHookSpecification.html

upvoted 1 times

...

spikeme

2 years ago

Agree A

upvoted 1 times

...

Bulti

2 years, 4 months ago

Answer is A. D is incorrect as there is no such lifecycle hooks on System Manager. I am just not sure how we could install Puppet using CodeDeploy. Looks like to me it is similar to installing CloudWatch Agent with CodeDeploy using an S3 deployment.

upvoted 1 times

...

saeidp

2 years, 5 months ago

Yes, A is correct

upvoted 1 times

...

luk3k0

2 years, 5 months ago

Selected Answer: A

Agree A

upvoted 1 times

...

flavins

2 years, 6 months ago

Selected Answer: D

Why run codedeploy twice?

upvoted 2 times

...

Goozian

2 years, 8 months ago

Selected Answer: A

it's ASG, and all management is being done through ASG; Other services have no idea about ASG operation ( Scale in/out, ..) by default. The only reasonable response will be A

upvoted 2 times

...

vagobago

2 years, 9 months ago

Selected Answer: A

A seems to be the right solution. Highest voted D seems suspect - there is no such thing like "Systems Manager instance stop lifecycle hook"

upvoted 5 times

...

hubekpeter

2 years, 9 months ago

Selected Answer: D

The most easy and straight forward solution. https://aws.amazon.com/blogs/infrastructure-and-automation/run-code-before-terminating-an-ec2-auto-scaling-instance/

upvoted 6 times

...