Skip to content

overview.md

Latest commit

 

History

History
216 lines (154 loc) · 5.38 KB

overview.md

File metadata and controls

216 lines (154 loc) · 5.38 KB

Deployment Overview

Warning

DISCLAIMER: This software is for educational and research purposes only. Trading involves significant risk of loss. Deploying bots to run on a schedule or for live trading is strictly at your own risk. The authors and contributors are not liable for any financial losses or damages incurred.

The trading bot system supports two deployment approaches:

  1. Kubernetes with Helm (Production) - Deploys PostgreSQL and bots as CronJobs
  2. Local Development - Run PostgreSQL locally, execute bots manually

Deployment Options

Option 1: Kubernetes with Helm (Production)

Deploy the entire system to Kubernetes, including PostgreSQL and all trading bots as scheduled CronJobs.

Benefits:

  • Automated scheduling via Kubernetes CronJobs
  • PostgreSQL deployed in-cluster
  • Centralized configuration via Helm
  • Production-ready with resource limits and monitoring

Prerequisites:

  • Kubernetes cluster (1.20+)
  • kubectl configured
  • Helm 3.x installed

Quick Deploy:

# 1. Create namespace
kubectl create namespace tradingbots-2025

# 2. Create secrets from .env file
kubectl create secret generic tradingbot-secrets \
  --from-env-file=.env \
  --namespace=tradingbots-2025

# 3. Deploy with Helm
helm upgrade --install tradingbots \
  ./helm/tradingbots \
  --create-namespace \
  --namespace tradingbots-2025

What Gets Deployed:

  • PostgreSQL instance (if postgresql.enabled: true in values.yaml)
  • Trading bot CronJobs (one per bot in values.yaml)
  • Visualization dashboard (if visualization.enabled: true)
  • All configured with secrets and environment variables

See Kubernetes Deployment and Helm Charts for detailed configuration.

Option 2: Local Development

Run PostgreSQL locally and execute bots manually with bot.run().

Benefits:

  • Fast iteration and testing
  • No Kubernetes cluster required
  • Easy debugging and development

Setup:

  1. Run PostgreSQL with Docker:
docker run -d --name postgres-tradingbot \
  -e POSTGRES_PASSWORD=yourpassword \
  -e POSTGRES_DB=tradingbot \
  -p 5432:5432 \
  postgres:17-alpine
  1. Set Environment Variable:
export POSTGRES_URI="postgresql://postgres:yourpassword@localhost:5432/tradingbot"
  1. Run Your Bot:
from tradingbot.utils.botclass import Bot

bot = MyBot()
bot.run()  # Executes one iteration

See the Quick Start Guide for complete local development workflow.

PostgreSQL Deployment

Kubernetes (via Helm)

PostgreSQL is automatically deployed when postgresql.enabled: true in helm/tradingbots/values.yaml:

postgresql:
  enabled: true
  image:
    repository: postgres
    tag: 17-alpine
  database: postgres
  service:
    name: psql-service
    port: 5432

The Helm chart creates:

  • PostgreSQL Deployment
  • PostgreSQL Service (accessible at psql-service:5432)
  • Persistent storage (configured in values.yaml)

Connection String: postgresql://postgres:${POSTGRES_PASSWORD}@psql-service:5432/postgres

Local (Docker)

For local development, use the Docker command shown in Option 2 above.

Bot Deployment

Kubernetes (CronJobs)

Bots are deployed as Kubernetes CronJobs with configurable schedules:

# helm/tradingbots/values.yaml
bots:
  - name: mybot
    schedule: "*/5 * * * 1-5"  # Every 5 minutes, Mon-Fri

Each bot runs automatically on its schedule. See Helm Charts for configuration details.

Local (Manual Execution)

Run bots manually with bot.run():

bot = MyBot()
bot.run()  # Execute one iteration

For scheduled execution locally, use cron or a task scheduler.

Configuration

Secrets Management

All secrets (database passwords, API keys) are stored in a single Kubernetes secret:

# Create .env file with:
# POSTGRES_PASSWORD=yourpassword
# POSTGRES_URI=postgresql://postgres:yourpassword@psql-service:5432/postgres
# OPENROUTER_API_KEY=yourkey
# BASIC_AUTH_PASSWORD=yourpassword

# Create secret
kubectl create secret generic tradingbot-secrets \
  --from-env-file=.env \
  --namespace=tradingbots-2025

Security: Never commit .env files to version control. Add to .gitignore.

Environment Variables

Bots receive environment variables from Kubernetes secrets:

env:
  - name: POSTGRES_URI
    valueFrom:
      secretKeyRef:
        name: tradingbot-secrets
        key: POSTGRES_URI

Monitoring

Kubernetes

View bot execution logs:

# List CronJobs
kubectl get cronjobs -n tradingbots-2025

# List recent jobs
kubectl get jobs -n tradingbots-2025

# View logs
kubectl logs -n tradingbots-2025 job/<job-name>

Local

Check database for run logs:

from tradingbot.utils.db import get_db_session, RunLog

with get_db_session() as session:
    logs = session.query(RunLog).filter(
        RunLog.bot_name == "MyBot"
    ).order_by(RunLog.timestamp.desc()).limit(10).all()
    
    for log in logs:
        print(f"{log.timestamp}: {log.success} - {log.result}")

Next Steps