ChromeOS Log Analyzer

A Python tool for parsing, analyzing, and visualizing ChromeOS logs. Generates interactive HTML reports with time-series charts, error markers, and log browsing capabilities.

Features

• Multiple Input Formats: Supports generated_logs directories, ZIP archives, and user feedback text files
• Auto-Detection: Automatically identifies input format
• Interactive Charts: Time-series visualization of CPU, memory, swap, and thermal data using Chart.js
• Error Detection: Identifies critical errors, warnings, and failures with timeline markers
• Log Browser: Searchable log viewer with keyword highlighting
• Self-Contained Output: Single HTML file with embedded CSS, JavaScript, and data
• Configurable: JSON-based configuration with CLI overrides

Installation

# Clone the repository
git clone https://github.com/user/chromeos_log_parser.git
cd chromeos_log_parser

# Create virtual environment (optional but recommended)
python3 -m venv venv
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

Usage

Basic Usage

# Analyze extracted generated_logs directory
python -m src.main -i ./generated_logs/ -o report.html

# Analyze user feedback text file
python -m src.main -i feedback.txt -o report.html

# Analyze ZIP archive directly
python -m src.main -i logs.zip -o report.html

# With verbose output
python -m src.main -i logs/ -o report.html --verbose

Command Line Options

-i, --input       Input path: directory, ZIP file, or text file (required)
-o, --output      Output HTML report path (default: chromeos_log_report.html)
-t, --type        Log type: auto, generated, or feedback (default: auto)
-v, --verbose     Enable verbose output
--max-points      Maximum chart data points (default: from config.json)
--max-duration    Maximum chart duration in minutes (default: from config.json)
--year            Reference year for timestamps (default: current year)
--config          Path to config.json file (default: auto-detect)

Configuration

Settings can be customized via config.json:

{
    "chart": {
        "max_duration_minutes": 10,
        "max_data_points": 5000
    },
    "parsing": {
        "time_gap_threshold_seconds": 5
    },
    "error_detection": {
        "default_marker_level": "none"
    }
}

CLI arguments override config file settings.

Generated Report

The HTML report includes:

1. System Info: ChromeOS version, board, and metadata
2. Error Summary: Categorized errors with search and filter capabilities
3. Time-Series Charts:
   • CPU usage (0-100%)
   • Memory usage
   • Swap usage
   • Thermal data (temperature zones)
4. Log Browser: Full log content with search and keyword highlighting
5. Interactive Features:
   • Chart zoom and pan
   • Error marker toggle (on/off)
   • Error level filter (none/error/critical/all)
   • Hover tooltips for error details

Supported Log Formats

Generated Logs Format

Directory structure from ChromeOS feedback:

generated_logs/
├── var_log/
│   ├── messages
│   ├── vmlog/
│   │   ├── vmlog.LATEST
│   │   └── vmlog.PREVIOUS
│   └── temp_logger/
│       └── temp_logger
├── lsb-release
└── etc/

User Feedback Format

Single text file with sections:

CHROMEOS_RELEASE_VERSION=15633.0.0
CHROMEOS_RELEASE_BOARD=volteer

vmlog.LATEST=<multiline>
---------- START ----------
time pgmajfault pgmajfault_f pgmajfault_a pswpin pswpout cpuusage cpufreq0 cpufreq1 cpufreq2 cpufreq3
[1027/151447] 0 0 0 0 0 0.02 700000 700000 698611 1263820
---------- END ----------

syslog=<multiline>
---------- START ----------
...log content...
---------- END ----------

Architecture

Input (ZIP/Directory/Text)
         │
         ▼
┌─────────────────┐
│  Parser Layer   │  ─── Normalization & Extraction
└─────────────────┘
         │
         ▼
┌─────────────────┐
│ Analyzer Layer  │  ─── Pattern Detection & Statistics
└─────────────────┘
         │
         ▼
┌─────────────────┐
│Visualizer Layer │  ─── Chart Generation & HTML Assembly
└─────────────────┘
         │
         ▼
   HTML Report (Single file)

See docs/specs/architecture.md for detailed architecture documentation.

Project Structure

chromeos_log_parser/
├── src/
│   ├── main.py                   # CLI entry point
│   ├── parsers/
│   │   ├── generated_logs_parser.py
│   │   ├── user_feedback_parser.py
│   │   ├── vmlog_parser.py
│   │   └── temp_logger_parser.py
│   ├── analyzers/
│   │   ├── error_detector.py
│   │   └── metrics_analyzer.py
│   └── visualizers/
│       ├── chart_generator.py
│       └── html_builder.py
├── tests/
│   ├── fixtures.py               # Sample log data
│   ├── test_parsers.py
│   └── test_analyzers.py
├── docs/
│   └── specs/                    # Specification documents
├── config.json                   # Default configuration
├── requirements.txt
└── README.md

Running Tests

# Run all tests
pytest tests/ -v

# Run specific test file
pytest tests/test_parsers.py -v

# Run with coverage
pytest tests/ --cov=src --cov-report=html

HTML Report Features

Chart Interactions

• Zoom: Mouse wheel or pinch gesture
• Pan: Click and drag
• Reset: Double-click to reset view
• Error Markers: Toggle visibility with dropdown selector

Search Features

• Error Summary: Filter errors by keyword
• Log Browser: Search within log content with match highlighting
• Navigation: Previous/Next buttons to jump between matches

Error Level Filter

• None: Hide all error markers on charts
• Error: Show ERROR and WARNING level markers
• Critical: Show only CRITICAL and FATAL markers
• All: Show all detected error markers

Error Patterns Detected

The analyzer detects the following error patterns in logs:

Level	Patterns
Critical	FATAL, CRASH, panic, OOM, Out of memory, killed, CRITICAL
Error	ERROR, FAILED, failure, timeout, refused, denied
Warning	WARNING, WARN

Performance

• Target Processing Time: <60 seconds for typical logs
• Input Size: Up to 500MB
• Output Size: <10MB HTML file
• Data Sampling: Automatic sampling for large datasets

License

MIT License