This repository contains Terraform modules, Kubernetes manifests, and Helm charts for deploying Heretek OpenClaw to various cloud providers and on-premises environments.

Supported Platforms

Platform	Status	Documentation
AWS	✅ Stable	AWS Guide
GCP	✅ Stable	GCP Guide
Azure	🚧 Beta	Azure Guide
Kubernetes	✅ Stable	Kubernetes Guide
Docker	✅ Stable	Docker Guide
Bare Metal	✅ Stable	Bare Metal Guide

Quick Start

Prerequisites

Terraform 1.5+
kubectl (for Kubernetes deployments)
Helm 3+ (for Helm deployments)
Cloud provider CLI (aws, gcloud, az)

AWS Deployment

cd terraform/aws

# Initialize Terraform
terraform init

# Configure variables
cp terraform.tfvars.example terraform.tfvars
# Edit terraform.tfvars with your settings

# Plan deployment
terraform plan -out=tfplan

# Apply deployment
terraform apply tfplan

# Output will show access URLs and credentials

GCP Deployment

cd terraform/gcp

# Initialize Terraform
terraform init

# Configure variables
cp terraform.tfvars.example terraform.tfvars

# Plan deployment
terraform plan -out=tfplan

# Apply deployment
terraform apply tfplan

Kubernetes Deployment

Using Helm (Recommended)

cd helm

# Add Helm repo (if published)
helm repo add heretek https://heretek.github.io/charts

# Install OpenClaw
helm install openclaw ./openclaw -f values.yaml

# Or from source
helm install openclaw ./openclaw \
  --namespace openclaw \
  --create-namespace \
  --set gateway.replicas=1

Using kubectl

cd kubernetes

# Apply base manifests
kubectl apply -f base/

# Apply overlays for your environment
kubectl apply -k overlays/production/

Terraform Modules

Available Modules

Module	Description	Location
`gateway`	OpenClaw Gateway deployment	`terraform/modules/gateway/`
`litellm`	LiteLLM Gateway deployment	`terraform/modules/litellm/`
`database`	PostgreSQL with pgvector	`terraform/modules/database/`
`cache`	Redis cache cluster	`terraform/modules/cache/`
`networking`	VPC, subnets, security groups	`terraform/modules/networking/`

Module Usage Example

module "openclaw_gateway" {
  source = "./modules/gateway"
  
  cluster_name    = "openclaw-prod"
  instance_type   = "m5.xlarge"
  min_capacity    = 1
  max_capacity    = 5
  
  vpc_id          = module.vpc.vpc_id
  subnet_ids      = module.vpc.private_subnet_ids
  
  database_url    = module.database.connection_string
  redis_url       = module.cache.endpoint
}

Helm Charts

Chart Values

Key	Default	Description
`gateway.replicas`	`1`	Number of Gateway replicas
`gateway.image.tag`	`latest`	Gateway image tag
`litellm.enabled`	`true`	Enable LiteLLM
`database.enabled`	`true`	Deploy PostgreSQL
`redis.enabled`	`true`	Deploy Redis
`monitoring.enabled`	`true`	Enable monitoring stack

Custom Values Example

# values.yaml
gateway:
  replicas: 3
  resources:
    requests:
      cpu: 500m
      memory: 1Gi
    limits:
      cpu: 2000m
      memory: 4Gi

litellm:
  models:
    - name: agent/steward
      model: openai/gpt-4o

database:
  storage: 100Gi
  backup:
    enabled: true
    schedule: "0 2 * * *"

Configuration

Environment Variables

Variable	Description	Default
`GATEWAY_URL`	Gateway endpoint	`http://localhost:18789`
`LITELLM_URL`	LiteLLM endpoint	`http://localhost:4000`
`POSTGRES_HOST`	PostgreSQL host	`localhost`
`POSTGRES_PORT`	PostgreSQL port	`5432`
`REDIS_HOST`	Redis host	`localhost`
`REDIS_PORT`	Redis port	`6379`

Terraform Variables

Variable	Description	Default
`cluster_name`	Cluster name	`openclaw`
`region`	Cloud region	`us-west-2`
`environment`	Environment name	`production`
`instance_type`	Instance type	`m5.xlarge`

CI/CD Integration

GitHub Actions

name: Deploy OpenClaw

on:
  push:
    branches: [main]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      
      - uses: hashicorp/setup-terraform@v3
      
      - run: terraform init
      - run: terraform plan -out=tfplan
      - run: terraform apply tfplan

Monitoring

Health Checks

# Check Gateway health
curl http://localhost:18789/health

# Check LiteLLM health
curl http://localhost:4000/health

# Check PostgreSQL
pg_isready -h localhost -p 5432

# Check Redis
redis-cli ping

Metrics

Prometheus metrics available at /metrics endpoint
Grafana dashboards included in monitoring stack
Custom alerts configured in monitoring/alerts/

Troubleshooting

Common Issues

Terraform Apply Fails

# Check state
terraform state list

# Refresh state
terraform refresh

# Import existing resources if needed
terraform import <resource> <id>

Kubernetes Pods Not Starting

# Check pod status
kubectl get pods -n openclaw

# View logs
kubectl logs -n openclaw <pod-name>

# Describe pod for events
kubectl describe pod -n openclaw <pod-name>

Documentation

Core - Gateway and agents
CLI - Deployment CLI
Dashboard - Health monitoring
Plugins - Plugin system
Docs - Documentation site

License

MIT

Support

Issues: https://github.com/heretek/heretek-openclaw-deploy/issues
Discussions: https://github.com/heretek/heretek-openclaw-deploy/discussions

🦞 The thought that never ends.

Languages

HCL 86.5%

PLpgSQL 7.5%

JavaScript 3.3%

Shell 1.5%

Smarty 1.2%

README.md

Heretek OpenClaw Deploy

Overview