CP_AUTOMATION/tests/load_tests

World-Class Load Testing Script - Complete Guide

📖 Table of Contents

1. What is This?
2. Prerequisites
3. Quick Start
4. Understanding the Flow
5. Command Line Arguments
6. Multi-Device Execution
7. Understanding Results
8. Troubleshooting
9. Best Practices
10. Advanced Usage

---

🎯 What is This?

This is a world-class load testing script that simulates multiple students completing the full assessment flow simultaneously. It's designed to:

• ✅ Test backend/server performance under load
• ✅ Verify system stability with concurrent users
• ✅ Provide transparent, real-time metrics
• ✅ Support multi-device execution (distributed load testing)
• ✅ Use only 100% verified, reliable automation flows

What Each Student Does

Complete Flow (9 Steps):

1. Login → Excel password first, fallback to Admin@123
2. Password Reset → If modal appears (smart detection)
3. Profile Fill → Complete to 100% if incomplete (smart detection)
4. Navigate to Assessments → Go to assessments page
5. Start Assessment → Click first available assessment
6. Select Domain → Click first unlocked domain
7. Answer All Questions → Answer all questions in domain (handles all 5 question types)
8. Submit Assessment → Submit when all questions answered
9. Feedback → Submit domain feedback if modal appears

Total Time: ~3-5 minutes per student (depending on number of questions)

---

📋 Prerequisites

Required

• ✅ Python 3.8+
• ✅ Virtual environment activated
• ✅ Chrome browser installed (or will be auto-managed)
• ✅ ChromeDriver (auto-managed by webdriver-manager)
• ✅ CSV file with student data
• ✅ Backend server running (localhost:3983 for local)

🚀 Quick Setup (Fresh System)

For a completely fresh system, run the setup script:

cd /home/tech4biz/work/CP_Front_Automation_Test
chmod +x setup_load_test_environment.sh
./setup_load_test_environment.sh

This script will:

• ✅ Check Python version (3.8+)
• ✅ Create virtual environment
• ✅ Install all dependencies from requirements.txt
• ✅ Check Chrome/ChromeDriver
• ✅ Validate project structure
• ✅ Verify function signature
• ✅ Create necessary directories

Manual Setup (if needed):

# 1. Create virtual environment
python3 -m venv venv

# 2. Activate it
source venv/bin/activate

# 3. Install dependencies
pip install -r requirements.txt

CSV File Format

Your CSV file must have these columns (case-insensitive):

• Student CPID or student_cpid or cpid or CPID (required)
• Password or password or PASSWORD (optional, will use Admin@123 if missing)
• First Name (optional, for display)
• Last Name (optional, for display)

Example CSV:

Student CPID,Password,First Name,Last Name
STU001,Pass123,John,Doe
STU002,Pass456,Jane,Smith

---

🚀 Quick Start

Step 0: Setup (First Time Only)

If this is a fresh system, run the setup script first:

cd /home/tech4biz/work/CP_Front_Automation_Test
./setup_load_test_environment.sh

If already set up, just activate the virtual environment:

cd /home/tech4biz/work/CP_Front_Automation_Test
source venv/bin/activate

Step 2: Validate Function Signature (Recommended)

python3 tests/load_tests/validate_function_signature.py

Expected Output: ✅ Function signature is valid!

Step 3: Run Your First Test (1 Student)

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students_with_passwords_2025-12-12T13-19-32.csv \
    --start 0 \
    --end 1 \
    --workers 1 \
    --headless \
    --metrics-interval 1

Step 4: Scale Up (10 Students)

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students_with_passwords_2025-12-12T13-19-32.csv \
    --start 0 \
    --end 10 \
    --workers 10 \
    --headless \
    --metrics-interval 5

Step 5: Full Load Test (100 Students)

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students_with_passwords_2025-12-12T13-19-32.csv \
    --start 0 \
    --end 100 \
    --workers 100 \
    --headless \
    --metrics-interval 10

---

🔍 Understanding the Flow

What Happens Behind the Scenes

1. Script starts → Loads students from CSV (with range filter)
2. For each student (in parallel):
   • Creates Chrome browser (headless or visible)
   • Logs in (smart password handling)
   • Handles password reset if needed
   • Completes profile if needed
   • Navigates to assessments
   • Starts first assessment
   • Selects first domain
   • Answers ALL questions
   • Submits assessment
   • Handles feedback
   • Closes browser
3. Metrics collected → Real-time performance tracking
4. Results saved → JSON report generated

Smart Features

• Smart Login: Tries Excel password first, falls back to Admin@123
• Smart Password Reset: Only resets if needed (checks current password state)
• Smart Profile Completion: Only completes if profile is incomplete
• Smart Question Answering: Handles all 5 question types automatically
• Smart Error Recovery: Retries on transient failures

---

⚙️ Command Line Arguments

Required Arguments

Argument	Description	Example
--csv	Path to CSV file	students_with_passwords_2025-12-12T13-19-32.csv

Optional Arguments

Argument	Description	Default	Example
--start	Start index (0-based, excluding header)	0	0, 100, 200
--end	End index (exclusive, None = all remaining)	None	100, 200, 500
--workers	Max concurrent workers	All students	10, 50, 100
--headless	Run in headless mode	True	(flag)
--visible	Run in visible mode (overrides headless)	False	(flag)
--metrics-interval	Print metrics every N students	10	5, 10, 20

Examples

Basic usage:

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students.csv \
    --start 0 \
    --end 100

With all options:

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students.csv \
    --start 0 \
    --end 100 \
    --workers 50 \
    --headless \
    --metrics-interval 10

Visible mode (for debugging):

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students.csv \
    --start 0 \
    --end 5 \
    --workers 5 \
    --visible

---

🌐 Multi-Device Execution

This script supports distributed load testing across multiple devices. Each device runs a different range of students.

Example: 500 Students on 5 Devices

Device 1 (Students 0-99):

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students_with_passwords_2025-12-12T13-19-32.csv \
    --start 0 \
    --end 100 \
    --workers 100 \
    --headless

Device 2 (Students 100-199):

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students_with_passwords_2025-12-12T13-19-32.csv \
    --start 100 \
    --end 200 \
    --workers 100 \
    --headless

Device 3 (Students 200-299):

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students_with_passwords_2025-12-12T13-19-32.csv \
    --start 200 \
    --end 300 \
    --workers 100 \
    --headless

Device 4 (Students 300-399):

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students_with_passwords_2025-12-12T13-19-32.csv \
    --start 300 \
    --end 400 \
    --workers 100 \
    --headless

Device 5 (Students 400-500):

python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students_with_passwords_2025-12-12T13-19-32.csv \
    --start 400 \
    --end 500 \
    --workers 100 \
    --headless

Range Calculation

• --start 0 --end 100 = Students at indices 0-99 (100 students)
• --start 100 --end 200 = Students at indices 100-199 (100 students)
• --start 0 --end 500 = Students at indices 0-499 (500 students)

Note: Index 0 is the first student (after header row in CSV)

---

📊 Understanding Results

Real-Time Metrics

During execution, you'll see metrics printed every N students:

📊 REAL-TIME METRICS
================================================================================
⏱️  Elapsed Time:        125.3s
✅ Completed:            50
❌ Failed:               2
📈 Success Rate:         96.2%
⚡ Rate:                 0.40 students/sec
⏳ Avg Duration:         245.6s
❓ Avg Questions:         12.3
📊 Total Questions:      615

📋 STEP METRICS:
   login                : 100.0% success, 2.1s avg
   password_reset       :  95.0% success, 3.5s avg
   profile_completion   :  90.0% success, 15.2s avg
   assessment           :  96.0% success, 220.5s avg
================================================================================

Final Summary

After completion, you'll see:

================================================================================
LOAD TEST SUMMARY: Complete Assessment Flow
================================================================================

📊 OVERALL METRICS
  Total Users:           100
  Successful:           95 (95.00%)
  Failed:               5
  Skipped:               0
  Total Duration:        1250.5 seconds

⏱️  PERFORMANCE METRICS
  Average Duration:      245.6 seconds
  Min Duration:          180.2 seconds
  Max Duration:          320.5 seconds
  Median (P50):          240.1 seconds
  95th Percentile (P95): 310.2 seconds
  99th Percentile (P99): 318.5 seconds

📄 PAGE METRICS
  Avg Page Load Time:    2.3 seconds
  Avg Scroll Time:       0.5 seconds
  Scroll Smooth Rate:    98.0%

JSON Report

Results are saved to:

reports/load_tests/load_test_Complete_Assessment_Flow_{N}users_{timestamp}.json

Report contains:

• Summary metrics (success rate, durations, percentiles)
• Individual student results
• Step-by-step completion status
• Error details (if any)
• Page load metrics

---

🔧 Troubleshooting

Issue: "No module named 'utils'"

Solution:

# Make sure you're in the project root
cd /home/tech4biz/work/CP_Front_Automation_Test
source venv/bin/activate

Issue: "TypeError: got multiple values for argument 'headless'"

Status: ✅ FIXED - This was resolved by adding user_id as first parameter

Verification:

python3 tests/load_tests/validate_function_signature.py

Issue: "No students loaded"

Causes:

• CSV path is incorrect
• --start and --end are invalid
• CSV doesn't have Student CPID column

Solution:

• Check CSV path is correct
• Verify --start < --end
• Ensure CSV has required columns

Issue: High Failure Rate

Possible Causes:

• Too many concurrent browsers (reduce --workers)
• System resources exhausted (RAM/CPU)
• Backend server overloaded
• Network issues

Solutions:

• Reduce --workers to 20-50
• Monitor system resources (htop, free -h)
• Check backend server logs
• Test with smaller batch first

Issue: "DevToolsActivePort file doesn't exist"

Cause: Too many Chrome instances trying to start simultaneously

Solution:

• Reduce --workers (try 20-50)
• Add delay between browser starts
• Use headless mode (--headless)

Issue: Slow Execution

Causes:

• Too many concurrent browsers
• System resources limited
• Network latency
• Backend server slow

Solutions:

• Use --headless mode (faster)
• Reduce --workers
• Check network connection
• Monitor backend performance

---

💡 Best Practices

1. Start Small, Scale Up

• ✅ Test with 1 student first
• ✅ Then 10 students
• ✅ Then 50 students
• ✅ Finally 100+ students

2. Use Headless Mode

• ✅ Always use --headless for load testing
• ✅ Much faster and more stable
• ✅ Lower resource usage
• ✅ Use --visible only for debugging

3. Monitor System Resources

# In another terminal, monitor:
watch -n 1 'ps aux | grep chrome | wc -l'  # Count Chrome processes
htop  # Monitor CPU/RAM
free -h  # Check memory

4. Validate Before Running

# Always validate signature first
python3 tests/load_tests/validate_function_signature.py

5. Use Appropriate Concurrency

• Small test (1-10 students): --workers = number of students
• Medium test (10-50 students): --workers = 20-50
• Large test (50-100 students): --workers = 50-100
• Very large test (100+ students): --workers = 100 (or use multi-device)

6. Multi-Device Strategy

• Divide students evenly across devices
• Each device runs 100-200 students
• All devices run simultaneously
• Combine results from all devices

7. Metrics Interval

• Short tests (< 50 students): --metrics-interval 5
• Medium tests (50-100 students): --metrics-interval 10
• Long tests (100+ students): --metrics-interval 20

---

🚀 Advanced Usage

Custom Range Testing

Test specific students:

# Students 50-75
python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students.csv \
    --start 50 \
    --end 75 \
    --workers 25

Resuming Failed Tests

If a test fails, you can resume by running the same command with adjusted --start:

# Original: --start 0 --end 100 (failed at student 50)
# Resume from student 50:
python3 tests/load_tests/test_generic_load_assessments.py \
    --csv students.csv \
    --start 50 \
    --end 100 \
    --workers 50

Performance Analysis

Analyze JSON report:

import json

with open('reports/load_tests/load_test_Complete_Assessment_Flow_100users_*.json') as f:
    data = json.load(f)
    
summary = data['summary']
print(f"Success Rate: {summary['success_rate']:.2f}%")
print(f"Avg Duration: {summary['avg_duration']:.2f}s")
print(f"P95 Duration: {summary['p95_duration']:.2f}s")

---

📝 Notes

Password Strategy

• Excel password is tried first (from CSV)
• Admin@123 is used as fallback
• Password reset only happens if needed (smart detection)

Assessment Flow

• Completes ONE domain (first unlocked domain)
• Answers ALL questions in that domain
• Handles all 5 question types automatically
• Submits when all questions answered

Error Handling

• Robust retry logic for transient failures
• Smart error recovery
• Comprehensive error logging
• Driver cleanup on errors

Resource Management

• Automatic browser cleanup
• Thread pool management
• Memory-efficient execution
• Progress tracking

---

✅ Verification Checklist

Before running a large load test:

• Virtual environment activated
• Function signature validated (validate_function_signature.py)
• CSV file exists and has correct format
• Backend server is running
• Tested with 1 student first
• System resources are adequate
• Using headless mode for load testing
• Appropriate --workers value set
• Monitoring system resources

---

🆘 Getting Help

Check These First:

1. Run validation script: python3 tests/load_tests/validate_function_signature.py
2. Test with 1 student first
3. Check system resources
4. Review error messages in JSON report
5. Check backend server logs

Common Questions:

Q: How many students can I run simultaneously? A: Depends on system resources. Start with 20-50, scale up based on performance.

Q: Can I run this on multiple machines? A: Yes! Use --start and --end to divide students across machines.

Q: How long does it take? A: ~3-5 minutes per student. 100 students = ~5-8 hours (depending on concurrency).

Q: What if a student fails? A: The script continues with other students. Check the JSON report for details.

Q: Can I pause and resume? A: Not automatically, but you can use --start to resume from a specific index.

---

📚 Related Files

• test_generic_load_assessments.py - Main load test script
• validate_function_signature.py - Signature validation script
• VERIFICATION_SUMMARY.md - Issue analysis and resolution
• LOAD_TEST_USAGE.md - Quick reference guide

---

Last Updated: 2025-12-12 Status: ✅ Production Ready - All Issues Resolved