Loki JSONL Log Ingestion

This project provides a complete pipeline to ingest .jsonl-formatted logs into Grafana Loki using Promtail, and visualize them in Grafana.

🛠️ Features

• Ingest .jsonl structured logs via custom Python script
• Pre-configured docker-compose setup for Loki, Promtail, and Grafana
• Real-time log visualization in Grafana Explore
• Supports ingestion of large datasets with rate limit management

📦 Project Structure

.
├── docker-compose.yml          # Loki, Promtail, Grafana services
├── loki-config.yaml            # Loki configuration
├── promtail-config.yaml        # Promtail configuration
├── upload_all_json.py         # Python uploader for JSONL files

🚀 Getting Started

1. Clone the Repository

git clone https://github.com/yourusername/loki-jsonl-log-ingestion.git
cd loki-jsonl-log-ingestion

2. Launch the Stack

sudo docker-compose up -d

This will start:

• Loki (default port: 3100)
• Promtail (default port: 9080)
• Grafana (default port: 3000)

3. Upload JSONL Logs to Loki

python3 upload_all_jsonl.py ./by_source/

⚠️ Note: Some files may exceed the default Loki ingestion limit (e.g., 4MB). You'll need to split or downsample large files or adjust Loki's limits in the configuration.

4. Access Grafana

• Visit http://localhost:3000
• Default credentials:
  • Username: admin
  • Password: admin (you'll be prompted to change it)

5. Explore Logs

• Go to Explore
• Select Loki as the data source
• Use queries like:

  {job="files"}
  

⚙️ Configuration Notes

• Promtail is configured to tail files in /var/log/*.log and ingest sample .jsonl logs
• You can customize promtail-config.yaml to change file targets or relabel log streams

🧩 Troubleshooting

• 500 ResourceExhausted: File exceeds maximum allowed size (default 4MB). Try chunking the data.
• 429 Rate limit exceeded: Too many log entries per second. Use time.sleep() or chunk uploads.
• Use curl http://localhost:3100/ready to verify Loki is ready.
• Use docker-compose logs to debug individual service logs.