Quiz Flow

Introduction

The application implements a cycle scheduling process for assigning questions to users based on their region. The following flowchart illustrates the high-level operations involved in this process:

Cycles Scheduling Process

flowchart TD
  subgraph Admin Actions
    A[Admin Adds New Cycle] -->|Creates Cycle| B[(Database: RegionalQuestionCycle)]
  end
  
  subgraph Queue Processor
    B --> C[/Queue: Schedule Next Cycle/]
    C --> D[[Processor: Schedule Next Cycle]]
    D -->|Check if Overlapping Cycle Exists| E{Cycle Overlap?}
    E -->|Yes| F[Skip Cycle Creation]
    E -->|No| G[Check if questions for a given region available]
    G -->|No| F[Skip Cycle Creation]
    G -->|Yes: Create next cycle| B
    G -->|Schedule job with delay| H[/Queue:Process Job Before Cycle End/]
  end

  subgraph Queue Retry Mechanism
    D --> I[Retry 3 times if error]
  end

  H -->|Wait until delay| D

Loading

The Bullboard interface is available at http://localhost:3000/api/v1/bullboard for monitoring queue activities.

Overview & Other Details

The goal was to design a dynamic question rotation system where questions are assigned to regions on a cycle basis, with configurable cycle durations. The system allows admins to add new questions and assign them to specific regions, ensuring flexibility in question management.

The architecture revolves around asynchronous job processing to manage question assignments and cycle scheduling in a robust and automated manner.

Key Features

1. Cycle Scheduling: Admins can manually create cycles or let the system schedule them based on default configurations. This auto-pilot mode ensures that cycles run continuously without admin intervention.
2. Question Management: Admins can keep adding new questions and assign them to regions dynamically. The system ensures that every region gets its own set of questions for each cycle.
3. BullMQ Queues: Jobs are queued for scheduling the next cycle, ensuring cycles are processed in a timely manner. Delayed job processing ensures that the next cycle is scheduled before the current one ends.
4. Retry and Backoff: Jobs automatically retry in case of failures with exponential backoff to ensure robust error handling.

Additional Details

• After running the last seeder (npm run seed:regional-question-cycle), you can monitor the BullMQ job queue on Bullboard to see jobs being scheduled.
• Admin Credentials:
  • Email: [email protected]
  • Password: Admin@123

Pros

• Scalable: Capable of handling large-scale user bases through background job queues.
• Auto-Pilot Scheduling: The system schedules cycles automatically, reducing admin workload.
• Fault-tolerant: Retry mechanisms ensure reliable job processing even in case of failure.
• Flexible Management: Admins can add new questions anytime and let the system handle scheduling.

Cons

• Processing Delays: Delays in job processing may cause slight lags.

Usage

Copy .env.sample file and rename it to .env and update necessary configs

Install packages

npm install
npm run build

Migrate database

npx prisma migrate dev

seed required data

npm run seed:regions
npm run seed:admin
npm run seed:questions
npm run seed:question-assignments
npm run seed:regional-question-cycle

Run the project

npm run start:prod

Swagger

The Swagger interface is available at http://localhost:3000/swagger