Professional Cloud Architect Exam QuestionsBrowse all questions from this exam

Professional Cloud Architect Exam - Question 13


Your customer is receiving reports that their recently updated Google App Engine application is taking approximately 30 seconds to load for some of their users.

This behavior was not reported before the update.

What strategy should you take?

Show Answer
Correct Answer: C

The most appropriate strategy in this scenario is to roll back to an earlier known good release to immediately mitigate the impact on users. Following the rollback, using Stackdriver Trace and Logging to diagnose the problem in a development, test, or staging environment is a sound approach to ensuring the issue is accurately pinpointed without impacting the live production environment. This method minimizes downtime for users and allows thorough investigation and resolution of the problem, preserving application stability while addressing the root cause.

Discussion

54 comments
Sign in to comment
TosO
Nov 22, 2019

C is the answer

MyPractice
Sep 20, 2024

Key word: This behavior was not reported before the update A - Not Correct as it was working before with same ISP B - New code update caused an issue- why to open support ticket C - I agree with C D - This requires downtime and live prod affected too

MyPractice
Dec 26, 2019

"then use Stackdriver Trace and Logging to diagnose the problem in a development/test/staging environment" they are NOT asking us to setup Dev/Text/Stage.. meaning the environment already exist and we have to use it

hafid
Jun 15, 2020

"then use Stackdriver Trace and Logging to diagnose the problem in a development/test/staging environment" this is not asking for set environment either, it just says to diagnose problem in other environment so C it is

hafid
Jun 15, 2020

"then use Stackdriver Trace and Logging to diagnose the problem in a development/test/staging environment" this is not asking for set environment either, it just says to diagnose problem in other environment so C it is

Eroc
Oct 24, 2019

More Ambiguity, I'm defaulting to a balance between frugality and likelihood. "A" would only be necessary if the ISP(Internet Service Provider) was causing the problem, most ISPs provide an error response faster than 30seconds. "B" would work, but Google Support agents are expensive... up from $50,000 annually. "C" is the best option as it will allow the GCP professional to find the root cause of the problem without increasing customer costs. "D" is the same as "C" but is testing in the exposed code based, "D" should only be chosen if the customer is not paying for other code bases already.

tartar
Aug 6, 2020

C is ok

nitinz
Mar 4, 2021

C is the best option.

clouddude
May 10, 2020

I'll go with C. A does not seem reasonable because the problem happened after the update and is likely not ISP-related. B does not seem reasonable even if Support could assist because that would depend on your support level and doesn't take advantage of the tools you have available. C seems a reasonable first approach as it gets production back quickly and takes advantage of the ability to provision a cloud staging environment immediatelyh. D seems feasible but not desirable because it runs the risk of interfering with users again and imposes a delay.

AD2AD4
May 28, 2020

Final Decision to go with Option C

Ziegler
May 29, 2020

C is very correct

Ekramy_ElnaggarOption: C
Nov 11, 2024

1. Prioritize User Experience: Rolling back to a stable version quickly minimizes user impact and restores the application to a functional state. This should be the immediate first step. 2. Controlled Environment: Diagnosing the issue in a development/test/staging environment allows you to investigate without affecting real users. You can reproduce the problem, gather data, and test potential solutions safely. 3. Powerful Diagnostic Tools: Stackdriver Trace helps you pinpoint performance bottlenecks by tracing requests across your application. Stackdriver Logging provides detailed logs to understand application behavior and identify errors.

Smart
Feb 9, 2020

A (Incorrect): Problem began with new version; B (Incorrect): Problem is about latency - not traffic restriction. D (Incorrect): CI/CD Pipeline encourages easy rollbacks and troubleshooting in Dev/Test environment. C is the most appropriate way to diagnose and troubleshoot.

Nirms
Jun 2, 2020

C is the correct answer

mlantonis
Jun 23, 2020

I'll go with C

cmfchong
Apr 10, 2021

I would choose D. Because Roll back to an earlier known good release initially, then install StackDriver for option C without install the updated app would not allow you to able to find the problems in the updated app. For option C it did not say it will then install the updated app.

minmin2020Option: C
Oct 13, 2022

C. Roll back to an earlier known good release initially, then use Stackdriver Trace and Logging to diagnose the problem in a development/test/staging environment A and B are not relevant D - no IT manager will ever allow re-deployment of erroneous code in production, even in a quiet period...!

Kiroo
May 2, 2023

I agree why not D, but in the past I faced issues only reproducible in prd, at that situation D was a possibility but usually yep C is for sure

JC0926
Mar 13, 2023

Option C is also a valid strategy in this scenario. Rolling back to an earlier known good release initially and using Stackdriver Trace and Logging to diagnose the problem in a development/test/staging environment can help diagnose the issue without impacting production users. However, the reason why option D may be a better approach is that it allows for investigation during a quieter period, which can reduce the impact of any issues that may occur during the investigation. Rolling back to a known good release and then pushing the release again at a quieter period can help to ensure that users are not impacted during the investigation.

AWS56
Nov 19, 2019

D is the answer

gh999l
Dec 26, 2020

C is the answer

AWS56
Jan 11, 2020

C is the answer

2g
Jan 30, 2020

answer: C

PRC
Apr 14, 2020

Agree with C...

gcp_aws
May 7, 2020

C is the answer

gfhbox0083
Jun 10, 2020

C, for sure. Roll back to an earlier known good release initially, then use Stackdriver Trace and Logging to diagnose the problem in a development/test/staging environment

RM07
Jul 9, 2020

C is the answer without any impact on user.

Joyjit_Deb
Feb 15, 2021

I would rather go with "B". Rolling back (as suggested in "C" & "D") without enough evidence or investigations could be not a good approach.

Ausias18
Mar 30, 2021

answer is C

lynx256
Mar 30, 2021

C is ok

victory108
May 18, 2021

C. Roll back to an earlier known good release initially, then use Stackdriver Trace and Logging to diagnose the problem in a development/test/staging environment

TotoroChina
Jun 29, 2021

I go with D. How can we recurrent the problem after rollback to a good former release? You should see nothing useful for the problem from Trace & Logging when you are running on a good release. C should be correct only when you have trace & logging always on for the period this problems happened, which is not mentioned in the question and not practical as the cost may surge if you are running a busy service.

Pime13Option: C
Jan 29, 2022

choose C

NircaOption: C
Apr 20, 2022

C is the answer.

alekonkoOption: C
Mar 23, 2023

C is the answer

FigVamOption: C
May 6, 2023

should be C

frankryuuOption: C
Jun 28, 2023

Although it sounds like the right answer to do network tracing in stg again, this may be a network pass-through related issue and it is felt that the problem may not be reproduced if not checked in a prod environment.

jrisl1991Option: C
Sep 24, 2023

I'm going for C. While D may be "better" in case this is an issue that only occurs in production, I think that keeping the disruption at minimum would be the best practice, which D would not really do. Plus, if the problem is load related, having this released at a quieter period may not surface the problem either.

gkdinesh
Sep 10, 2020

Agree with option C

AshokC
Sep 14, 2020

C is more meaningful.

subhala
Sep 26, 2020

seems like a C. don't want impact customers before it is tested well.

nimso
Oct 26, 2020

I'll go with C

BobBui
Jan 28, 2021

I go with C

rmout
Feb 24, 2021

application is taking approximately 30 seconds to load for "some of their users" not All ! So in my opinion , we would want to get hold of the network and data flow log evidences before rollingback. Testing in lower environments might not always work unless you have like to like env , and might not be even able to replicate the issue. Guess "B" would be the right option

un
May 10, 2021

i will go with C

aviratna
Jun 26, 2021

C: Correct answer App engine gives flexibility to roll back to previous version. Priority should be restore the services to working state. And trace the issue using Stackdriver where the logs are already captured from previous failed service.

DreamerK
Jul 25, 2021

Why C is correct is there should be an assumption made by the author that new release will be deployed in the dev/test/stage environment. Otherwise C doesn't make any sense here.

amxexam
Aug 24, 2021

C is the ans For all those calling for D , the option dont tell they will install stackdriver, but will use it to analysie the logs. Then deploying back the faulty code without fixing does not make sences.

sandipk91
Aug 30, 2021

C is the only answer

haroldbenites
Dec 3, 2021

Go for C

vincy2202
Dec 24, 2021

C is the correct answer

HeyBuddy95Option: C
Mar 28, 2022

Answer is C

pfilourencoOption: C
Jun 23, 2022

The correct answer is c.

Amit_archOption: D
Sep 10, 2022

How come everyone is agreeing to C!! In option C after rollback, the investigation will happen only on the earlier good release. Whereas in option D, all the troubleshooting will happen on current/problematic build. Option D should be the right option as it resolves the issue in short term and provides room for further investigation without downtime.

BiddlyBdoyng
Sep 25, 2022

Option C is investigating the bad build in test. The problem with option D is it is user impacting. Always best to attempt to find the problem in a test environment first. D could end-up being an option of last resort if all attempts to diagnose in test fail but I doubt any business person would be happy with D as it impacts service.

zr79
Oct 17, 2022

you want to minimize the business loose, best option is to rollback and use stack-driver to diagnose the issue

holerina
Sep 20, 2022

correct answer is C use the standard practise

AzureDP900
Oct 17, 2022

C is perfect to troubleshoot latency issues with app

Mahmoud_EOption: C
Oct 22, 2022

C is the correct answer

meguminOption: C
Nov 5, 2022

ok for C

frankryuu
Jun 28, 2023

Although it sounds like the right answer to do network tracing in stg again, this may be a network pass-through related issue and it is felt that the problem may not be reproduced if not checked in a prod environment.

AdityaGuptaOption: C
Oct 4, 2023

Your customer is receiving reports that their recently updated Google App Engine application is taking approximately 30 seconds to load for some of their users. This behavior was not reported before the update. What strategy should you take? Here the application (our code) is updated and only some users are facing lantecy (Cloud Trace) issue. The issue is not with ISP (A), Not an issue with Google (B). Rollback must be done as mitigation, but testing should be done in Non-Prod environments (C), not on prod environment (D). Hence C is correct answer.

hzaouiOption: C
Jan 11, 2024

C is correct

QileeOption: C
May 4, 2025

C provides all options to provide and backup this situation.