Exam Professional Cloud Architect All Questions

View all questions & answers for the Professional Cloud Architect exam

Exam Professional Cloud Architect topic 1 question 127 discussion

Actual exam question from Google's Professional Cloud Architect

Question #: 127
Topic #: 1

[All Professional Cloud Architect Questions]

Your operations team has asked you to help diagnose a performance issue in a production application that runs on Compute Engine. The application is dropping requests that reach it when under heavy load. The process list for affected instances shows a single application process that is consuming all available CPU, and autoscaling has reached the upper limit of instances. There is no abnormal load on any other related systems, including the database. You want to allow production traffic to be served again as quickly as possible. Which action should you recommend?

A. Change the autoscaling metric to agent.googleapis.com/memory/percent_used.
B. Restart the affected instances on a staggered schedule.
C. SSH to each instance and restart the application process.
D. Increase the maximum number of instances in the autoscaling group.

Show Suggested Answer

Suggested Answer: D 🗳️

by XDevX at July 1, 2021, 5:27 a.m.

Comments

Submit Cancel

TotoroChina

Highly Voted 3 years, 6 months ago

Answer should be D. I doubt it is intended to provide wrong answer.

upvoted 45 times

poseidon24

3 years, 5 months ago

Agree. Cannot be A), since changing the metric used for autoscaling will not solve the issue, the CPU is already over utilized, hence the unique "workaround" meanwhile the application causing the issue is fixed (connection leaks, infinite loops, etc.) is to allow introducing new nodes/workers/VMs.

upvoted 9 times

...

victorlie

3 years, 4 months ago

why almost all answers are wrong?

upvoted 15 times

zr79

2 years, 2 months ago

to prevent us from memorizing the answers and hopefully, the site can not be shut down

upvoted 10 times

Action

1 year ago

Yes I always consider this to be the reason behind so many obviously wrong answers too... but who knows

upvoted 2 times

...

MamthaSJ

Highly Voted 3 years, 5 months ago

Answer is D

upvoted 12 times

...

AWS_Sam

Most Recent 1 year ago

The question is not asking for a permanent solution to the problem, it is asking what to do to have the production traffic to be served again as quickly as possible. Therefore, the best answer is D.

upvoted 4 times

...

odacir

1 year, 1 month ago

Selected Answer: D

Answer D is correct. Prioritize the availability on production environment.

upvoted 2 times

...

gary_cooper

1 year, 5 months ago

Selected Answer: D

Increase the maximum number of instances in the autoscaling group

upvoted 1 times

...

red_panda

1 year, 6 months ago

Selected Answer: D

Answer D is correct. In order to prioritize the availability on production environment (as per question), first we need to increase the number of max instances in the instance group, then, for sure we can investigate and restart application process. Be careful, often the answer is in the question

upvoted 1 times

...

grejao

1 year, 9 months ago

I choose for A, but D is the best choice. The trick is: "process that is consuming all available CPU" and "autoscaling has reached the upper limit of instances" If the process is consuming all available CPU, we need to reconfigure our metrics for best tresholds (Option A) AND if the autoscaling reached the upper limit of instances, so we need to increase this limit (Option D), BUT, after reached the upper limit of instances, it doesn't matter the tresholds, the process will consume all resources that have available. So, option D is the best option.

upvoted 9 times

Sur_Nikki

1 year, 7 months ago

Loved the way you made us travel through the roots of the question

upvoted 2 times

...

CGS22

1 year, 10 months ago

Selected Answer: D

The application is dropping requests because the available CPU is exhausted. Autoscaling has reached the upper limit of instances, so it cannot increase the number of instances to meet the demand. The best way to allow production traffic to be served again is to increase the maximum number of instances in the autoscaling group. This will allow autoscaling to increase the number of instances to meet the demand without exhausting the available CPU. Restarting the affected instances or SSHing to each instance and restarting the application process will not solve the problem because the root cause is that there are not enough instances to meet the demand.

upvoted 6 times

...

omermahgoub

2 years ago

The correct answer is D: Increase the maximum number of instances in the autoscaling group. If the application is dropping requests under heavy load and the process list for affected instances shows a single application process consuming all available CPU, increasing the maximum number of instances in the autoscaling group may help to alleviate the performance issue. By adding more instances to the group, you can distribute the load across multiple instances, which should help to reduce the strain on any single instance. This will allow production traffic to be served again more quickly.

upvoted 7 times

omermahgoub

2 years ago

Option A is incorrect because changing the autoscaling metric to agent.googleapis.com/memory/percent_used will not address the root cause of the performance issue. The issue is related to CPU utilization, not memory usage. Option B is incorrect because restarting the affected instances on a staggered schedule will not address the root cause of the performance issue. It may provide temporary relief, but the issue is likely to recur once the instances are under heavy load again. Option C is incorrect because restarting the application process on each instance will not address the root cause of the performance issue. It may provide temporary relief, but the issue is likely to recur once the instances are under heavy load again. Increasing the maximum number of instances in the autoscaling group is a more effective solution in this case.

upvoted 4 times

...

SureshbabuK

2 years ago

Selected Answer: B

Given the is not no abnormal load, autoscaling will is not required, restarting should kill the single application process consuming excess CPU

upvoted 1 times

...

surajkrishnamurthy

2 years ago

Selected Answer: D

D is the correct answer

upvoted 1 times

...

sanait100

2 years, 1 month ago

The keyword is "You want to allow production traffic to be served again as quickly as possible" so D should be the only answer so as to resume production traffic and then you can do a root cause analysis and take further action depending upon the findings.

upvoted 2 times

...

megumin

2 years, 1 month ago

Selected Answer: D

D is ok

upvoted 1 times

...

minmin2020

2 years, 2 months ago

Selected Answer: D

It all depends on how you want to troubleshoot the issue. Do you want to check the application before or after increasing the max number of instances in the scaling group. I guess in real life people will ask for an increase in the max number of instances and if the application process continues to consume all the CPU then they will probably stop/restart the app. D is the only sensible option. A is not an option B you could restart but you dont know if that will fix the issue C SSH assumes unix vm's (?)....!

upvoted 3 times

AzureDP900

2 years, 2 months ago

autoscaling has reached the upper limit of instances. There is no abnormal load on any other related systems, including the database. This so junk question, only D seems viable option.

upvoted 2 times

...