Skip to content

Deployment Overview

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

  2. Set Environment Variable: 

    export POSTGRES_URI="postgresql://postgres:yourpassword@localhost:5432/tradingbot"

  3. 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