CP_AUTOMATION/scripts/LOAD_TEST_README.md

# Load Testing Script - End-to-End Assessment Flow

## Overview

This script performs **load testing** by running multiple students through the complete end-to-end flow simultaneously:

1. ✅ **Login** (with smart password handling)
2. ✅ **Password Reset** (if needed)
3. ✅ **Profile Completion** (if needed)
4. ✅ **Complete ONE Domain Assessment** (100% reliable flow)

## Prerequisites

- ✅ **100% Working Flow**: Only uses flows we're 100% confident about
- ✅ **Single Domain Assessment**: Uses the most reliable assessment flow
- ✅ **All 5 Question Types**: Handles Multiple Choice, True/False, Rating Scale, Open Ended, Matrix
- ✅ **Error Handling**: Robust error handling and recovery
- ✅ **Randomized Waits**: Realistic timing for load testing

## Usage

### Basic Usage

```bash
python scripts/load_test_e2e_assessment.py --students 10 --csv students.csv
```

### With Custom Concurrency

```bash
python scripts/load_test_e2e_assessment.py --students 50 --csv students.csv --concurrent 10
```

### Visible Mode (for debugging)

```bash
python scripts/load_test_e2e_assessment.py --students 5 --csv students.csv --no-headless
```

## Arguments

| Argument | Required | Default | Description |
|----------|----------|---------|-------------|
| `--students` | ✅ Yes | - | Number of students to test |
| `--csv` | ✅ Yes | - | Path to CSV file with student data |
| `--concurrent` | ❌ No | 5 | Number of concurrent students |
| `--headless` | ❌ No | True | Run in headless mode |
| `--no-headless` | ❌ No | - | Run in visible mode (for debugging) |

## CSV Format

The CSV file should contain student data with at least:
- `cpid` or `CPID` column (or first column)
- `password` column (for initial password)
- Other student data (first_name, last_name, etc.) for profile completion

Example:
```csv
cpid,password,first_name,last_name,date_of_birth
SC309TB0284,VXa$Ai5kj4rV,John,Doe,2008-01-15
BAR210A010D,Admin@123,Jane,Smith,2007-05-20
```

## What It Does

### For Each Student:

1. **Setup WebDriver** - Creates a new browser instance
2. **Login** - Logs in with smart password handling (tries Excel password, then Admin@123)
3. **Password Reset** - Resets password if needed (only if not already reset)
4. **Profile Completion** - Completes profile to 100% if needed (only if incomplete)
5. **Navigate to Assessments** - Goes to assessments page
6. **Start Assessment** - Starts the first available assessment
7. **Select Domain** - Selects the first unlocked domain
8. **Answer All Questions** - Answers all questions in the domain (all 5 types)
9. **Submit Domain** - Submits the domain assessment
10. **Provide Feedback** - Provides domain feedback
11. **Cleanup** - Closes browser instance

## Output

The script provides:
- ✅ Real-time progress updates
- ✅ Success/failure status for each student
- ✅ Duration tracking
- ✅ Error reporting
- ✅ Summary statistics

Example output:
```
🚀 Starting flow for: SC309TB0284
✅ Completed: SC309TB0284 (success) - 245.32s
🚀 Starting flow for: BAR210A010D
✅ Completed: BAR210A010D (success) - 198.76s

================================================================================
LOAD TEST SUMMARY
================================================================================
✅ Success: 8/10
❌ Failed: 2/10
⏭️  Skipped: 0/10
⏱️  Total Duration: 1205.45s
📊 Average per Student: 120.54s
```

## Performance Considerations

### Recommended Settings

- **Small Load (1-10 students)**: `--concurrent 3-5`
- **Medium Load (10-50 students)**: `--concurrent 5-10`
- **Large Load (50+ students)**: `--concurrent 10-20`

### System Requirements

- **RAM**: ~500MB per concurrent student
- **CPU**: Multi-core recommended for concurrent execution
- **Network**: Stable connection to backend server

## Error Handling

The script handles:
- ✅ Login failures
- ✅ Password reset failures
- ✅ Profile completion failures
- ✅ Assessment navigation failures
- ✅ Question answering failures
- ✅ Network timeouts
- ✅ Browser crashes

All errors are logged and reported in the summary.

## Limitations

- ⚠️ **Single Domain Only**: Completes only ONE domain per student (the most reliable flow)
- ⚠️ **No Multi-Domain**: Does not complete all 6 domains (not yet 100% reliable)
- ⚠️ **No Final Feedback**: Does not handle final feedback after all domains (not yet 100% reliable)

## Future Enhancements

- [ ] Multi-domain completion (when 100% reliable)
- [ ] Final feedback handling (when 100% reliable)
- [ ] Performance metrics collection
- [ ] Backend response time tracking
- [ ] Database query monitoring
- [ ] Resource usage monitoring

## Troubleshooting

### "No students loaded from CSV"
- Check CSV file path
- Verify CSV has `cpid` or `CPID` column
- Check file encoding (should be UTF-8)

### "Driver setup failed"
- Ensure Chrome browser is installed
- Check ChromeDriver is in PATH
- Verify Chrome version matches ChromeDriver version

### "No assessments available"
- Ensure student profile is 100% complete
- Check if assessments are assigned to students
- Verify environment (local vs live)

### High failure rate
- Reduce `--concurrent` value
- Check backend server capacity
- Verify network stability
- Check system resources (RAM, CPU)

## Example Commands

### Test with 5 students (debugging)
```bash
python scripts/load_test_e2e_assessment.py --students 5 --csv students.csv --no-headless
```

### Test with 20 students (medium load)
```bash
python scripts/load_test_e2e_assessment.py --students 20 --csv students.csv --concurrent 5
```

### Test with 100 students (large load)
```bash
python scripts/load_test_e2e_assessment.py --students 100 --csv students.csv --concurrent 10
```

## Notes

- ✅ **100% Reliable Flow**: Only uses flows we're 100% confident about
- ✅ **Single Domain**: Completes one domain assessment per student
- ✅ **All Question Types**: Handles all 5 question types automatically
- ✅ **Error Recovery**: Robust error handling and recovery mechanisms
- ✅ **Realistic Timing**: Uses randomized waits for realistic behavior

---

**Status**: ✅ **READY FOR USE** - All flows are 100% reliable and tested