Exam Professional Cloud Architect topic 1 question 4 discussion

It's A. AppEngine spins up new containers automatically according to the load. During peak traffic, HTTP requests originated by the same user could be served by different containers. Given that the variable `sessions` is recreated for each container, it might store different data. The problem here is that this Flask app is stateful. The `sessions` variable is the state of this app. And stateful variables in AppEngine / Cloud Run / Cloud Functions are problematic. A solution would be to store the session in some database (e.g. Firestore, Memorystore) and retrieve it from there. This way the app would fetch the session from a single place and would be stateless.

A is correct

Answer is A. The session variable is local to just a single instance. Here's why: * App Engine Instances: Google App Engine automatically scales your application by creating multiple instances to handle incoming traffic. Each instance runs independently and has its own memory space. * Session Variable: In the provided code, the `sessions` variable is a simple Python dictionary stored in the memory of each instance. This means each instance has its own copy of the `sessions` data. * The Problem: When a user logs in, their viewed articles are stored in the `sessions` variable of the instance that handled their request. If subsequent requests from the same user are routed to a different instance, that instance won't have any record of the previously viewed articles, causing them to be displayed again.

A The most likely cause of the issue described in the code is that the session variable is local to just a single instance. In this code, the session variable is defined as a local dictionary within the Flask application. This means that it is not shared across different instances of the application and will not be persisted between requests. As a result, when the application is running on multiple instances, each instance will have its own local copy of the session variable, and users may see news articles that they have already viewed on other instances.

The most likely cause of the reported issue is that the session variable is local to just a single instance. In the code provided, the sessions variable is a dictionary that stores the viewed news articles for each user. However, this variable is only stored in memory on the instance that handles the request, and it is not shared between instances. Therefore, when a new request is handled by a different instance, it will not have access to the same session data, and the user may see previously viewed news articles. To solve this problem, a shared session management system should be used that can be accessed by all instances. Google App Engine provides a few options for session management, such as using Memcache or Cloud Datastore to store the session data. By using a shared session management system, all instances can access the same session data, and users will not see previously viewed news articles.

A.The sessions dictionary is used to store user-specific data, such as which news articles have been viewed. This dictionary is created as an in-memory variable within the Flask app. In a cloud environment, like Google App Engine, the application may be running on multiple instances, especially during peak loads. Since the sessions dictionary is stored in memory, it is local to each instance. This means that if a user is routed to a different instance (due to load balancing), their session data will not be available on that new instance, causing the application to serve news articles they’ve already seen.

A is the right answer

Where in the code does it show that the session variable is local to just a single instance?

A is correct

stateful variable should be in firestore (redis).

I agree with ackdbd

A is the correct answer app becoming stateful and it should not be sin case of app engine, cloud run and functions

A. The session variable is local to just a single instance The others are not relevant

Thank you that was nice explanation

Where was this presented in the GCP Architecture training & labs?

vote A - rule out C,D not thing to do with problem - rule out B, Q is not mention datastore

A of course, it's just a code pb here