CodeGenie Deployment Guide

Local Deployment
Docker Deployment
Kubernetes Deployment
AWS Deployment
Production Best Practices

Local Deployment

Quick Start

# Clone repository
git clone https://github.com/your-org/codegenie.git
cd codegenie

# Run installation script
./scripts/install.sh

The installation script will:

Check Python and Ollama installation
Install required models
Create virtual environment
Install CodeGenie
Create default configuration

Manual Installation

# Create virtual environment
python3 -m venv venv
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Install CodeGenie
pip install -e .

# Start Ollama
ollama serve &

# Pull models
ollama pull llama3.1:8b
ollama pull codellama:7b

# Start CodeGenie
codegenie

Docker Deployment

Using Docker Compose (Recommended)

# Clone repository
git clone https://github.com/your-org/codegenie.git
cd codegenie

# Start services
docker-compose up -d

# View logs
docker-compose logs -f

# Stop services
docker-compose down

Using Docker Directly

# Build image
docker build -t codegenie:latest .

# Run container
docker run -d \
  --name codegenie \
  -p 8000:8000 \
  -p 11434:11434 \
  -v $(pwd)/projects:/workspace \
  -v codegenie-config:/root/.config/codegenie \
  -v codegenie-cache:/root/.cache/codegenie \
  -v ollama-models:/root/.ollama \
  -e OLLAMA_MODELS="llama3.1:8b,codellama:7b" \
  codegenie:latest

# View logs
docker logs -f codegenie

# Stop container
docker stop codegenie

Docker Configuration

Create a .env file:

# Environment
CODEGENIE_ENV=production

# Ollama
OLLAMA_HOST=http://localhost:11434
OLLAMA_MODELS=llama3.1:8b,codellama:7b

# Database
POSTGRES_HOST=postgres
POSTGRES_DB=codegenie
POSTGRES_USER=codegenie
POSTGRES_PASSWORD=secure_password

# Redis
REDIS_HOST=redis
REDIS_PORT=6379

# Resources
MAX_MEMORY=16G
MAX_CPU=4

Kubernetes Deployment

Prerequisites

Kubernetes cluster (1.20+)
kubectl configured
Persistent storage provisioner
At least 100GB storage for models

Deploy to Kubernetes

# Create namespace
kubectl create namespace codegenie

# Create secrets
kubectl create secret generic codegenie-secrets \
  --from-literal=postgres-password=secure_password \
  -n codegenie

# Apply deployment
kubectl apply -f deploy/kubernetes/deployment.yaml -n codegenie

# Check status
kubectl get pods -n codegenie
kubectl get services -n codegenie

# View logs
kubectl logs -f deployment/codegenie -n codegenie

Scale Deployment

# Scale to 3 replicas
kubectl scale deployment codegenie --replicas=3 -n codegenie

# Auto-scaling
kubectl autoscale deployment codegenie \
  --min=2 \
  --max=10 \
  --cpu-percent=70 \
  -n codegenie

Update Deployment

# Update image
kubectl set image deployment/codegenie \
  codegenie=codegenie:v0.3.0 \
  -n codegenie

# Rollback if needed
kubectl rollout undo deployment/codegenie -n codegenie

# Check rollout status
kubectl rollout status deployment/codegenie -n codegenie

AWS Deployment

Using CloudFormation

# Deploy stack
aws cloudformation create-stack \
  --stack-name codegenie \
  --template-body file://deploy/aws/cloudformation.yaml \
  --parameters \
    ParameterKey=InstanceType,ParameterValue=g4dn.xlarge \
    ParameterKey=KeyName,ParameterValue=your-key-pair \
    ParameterKey=VpcId,ParameterValue=vpc-xxxxx \
    ParameterKey=SubnetId,ParameterValue=subnet-xxxxx \
  --capabilities CAPABILITY_IAM

# Check status
aws cloudformation describe-stacks \
  --stack-name codegenie \
  --query 'Stacks[0].StackStatus'

# Get outputs
aws cloudformation describe-stacks \
  --stack-name codegenie \
  --query 'Stacks[0].Outputs'

Using EC2 Directly

# Launch EC2 instance
aws ec2 run-instances \
  --image-id ami-xxxxx \
  --instance-type g4dn.xlarge \
  --key-name your-key-pair \
  --security-group-ids sg-xxxxx \
  --subnet-id subnet-xxxxx \
  --user-data file://scripts/ec2-userdata.sh

# SSH to instance
ssh -i your-key.pem ubuntu@<instance-ip>

# Install CodeGenie
curl -fsSL https://raw.githubusercontent.com/your-org/codegenie/main/scripts/install.sh | bash

Using ECS

# Create ECS cluster
aws ecs create-cluster --cluster-name codegenie

# Register task definition
aws ecs register-task-definition \
  --cli-input-json file://deploy/aws/ecs-task-definition.json

# Create service
aws ecs create-service \
  --cluster codegenie \
  --service-name codegenie-service \
  --task-definition codegenie:1 \
  --desired-count 2 \
  --launch-type FARGATE

Production Best Practices

Security

Use HTTPS

# Generate SSL certificate
certbot certonly --standalone -d codegenie.example.com

# Configure nginx
server {
    listen 443 ssl;
    server_name codegenie.example.com;
    
    ssl_certificate /etc/letsencrypt/live/codegenie.example.com/fullchain.pem;
    ssl_certificate_key /etc/letsencrypt/live/codegenie.example.com/privkey.pem;
    
    location / {
        proxy_pass http://localhost:8000;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
    }
}

Enable Authentication

# config.yaml
security:
  authentication:
    enabled: true
    type: "jwt"
    secret_key: "${JWT_SECRET}"
  
  rate_limiting:
    enabled: true
    requests_per_minute: 60

Use Secrets Management

# AWS Secrets Manager
aws secretsmanager create-secret \
  --name codegenie/postgres-password \
  --secret-string "secure_password"

# Kubernetes Secrets
kubectl create secret generic codegenie-secrets \
  --from-literal=postgres-password=secure_password

Performance

Enable Caching

cache:
  enabled: true
  backend: "redis"
  redis_url: "redis://localhost:6379"
  ttl: 3600

Use GPU Acceleration

models:
  use_gpu: true
  gpu_memory_fraction: 0.8

Configure Resource Limits

resources:
  max_memory: "16GB"
  max_cpu: 4
  max_concurrent_tasks: 10

Monitoring

Enable Logging

logging:
  level: "INFO"
  format: "json"
  outputs:
    - type: "file"
      path: "/var/log/codegenie/app.log"
    - type: "cloudwatch"
      log_group: "/aws/codegenie"

Configure Metrics

monitoring:
  enabled: true
  prometheus:
    enabled: true
    port: 9090
  cloudwatch:
    enabled: true
    namespace: "CodeGenie"

Set Up Alerts

alerts:
  - name: "high_memory"
    condition: "memory_usage > 90%"
    action: "email"
    recipients: ["admin@example.com"]
  
  - name: "high_error_rate"
    condition: "error_rate > 5%"
    action: "slack"
    webhook: "https://hooks.slack.com/..."

Backup and Recovery

Automated Backups

# Create backup script
cat > /usr/local/bin/backup-codegenie.sh << 'EOF'
#!/bin/bash
BACKUP_DIR="/backups/codegenie/$(date +%Y%m%d)"
mkdir -p "$BACKUP_DIR"

# Backup config
cp -r ~/.config/codegenie "$BACKUP_DIR/config"

# Backup sessions
cp -r ~/.codegenie/sessions "$BACKUP_DIR/sessions"

# Backup database
pg_dump codegenie > "$BACKUP_DIR/database.sql"

# Upload to S3
aws s3 sync "$BACKUP_DIR" s3://my-backups/codegenie/$(date +%Y%m%d)
EOF

chmod +x /usr/local/bin/backup-codegenie.sh

# Schedule with cron
echo "0 2 * * * /usr/local/bin/backup-codegenie.sh" | crontab -

Disaster Recovery

# Restore from backup
./scripts/restore.sh /backups/codegenie/20240101

High Availability

Load Balancing

upstream codegenie {
    least_conn;
    server codegenie-1:8000;
    server codegenie-2:8000;
    server codegenie-3:8000;
}

server {
    listen 80;
    location / {
        proxy_pass http://codegenie;
    }
}

Database Replication

database:
  primary:
    host: "postgres-primary"
    port: 5432
  replicas:
    - host: "postgres-replica-1"
      port: 5432
    - host: "postgres-replica-2"
      port: 5432

Updates and Maintenance

Zero-Downtime Updates

# Rolling update
kubectl set image deployment/codegenie \
  codegenie=codegenie:v0.3.0 \
  --record

# Blue-green deployment
kubectl apply -f deploy/kubernetes/deployment-green.yaml
# Test green deployment
kubectl delete -f deploy/kubernetes/deployment-blue.yaml

Maintenance Mode

# Enable maintenance mode
codegenie maintenance enable

# Perform updates
./scripts/update.sh

# Disable maintenance mode
codegenie maintenance disable

Troubleshooting Deployment

Common Issues

Out of Memory

# Check memory usage
docker stats codegenie

# Increase memory limit
docker update --memory 16g codegenie

Model Download Failures

# Manual model download
docker exec -it codegenie ollama pull llama3.1:8b

Connection Issues

# Check Ollama service
docker exec -it codegenie curl http://localhost:11434/api/tags

# Check logs
docker logs codegenie

Health Checks

# API health
curl http://localhost:8000/health

# Ollama health
curl http://localhost:11434/api/tags

# Database health
docker exec -it codegenie-postgres pg_isready

Support

For deployment issues:

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

CodeGenie Deployment Guide

Table of Contents

Local Deployment

Quick Start

Manual Installation

Docker Deployment

Using Docker Compose (Recommended)

Using Docker Directly

Docker Configuration

Kubernetes Deployment

Prerequisites

Deploy to Kubernetes

Scale Deployment

Update Deployment

AWS Deployment

Using CloudFormation

Using EC2 Directly

Using ECS

Production Best Practices

Security

Performance

Monitoring

Backup and Recovery

High Availability

Updates and Maintenance

Troubleshooting Deployment

Common Issues

Health Checks

Support

FilesExpand file tree

DEPLOYMENT.md

Latest commit

History

DEPLOYMENT.md

File metadata and controls

CodeGenie Deployment Guide

Table of Contents

Local Deployment

Quick Start

Manual Installation

Docker Deployment

Using Docker Compose (Recommended)

Using Docker Directly

Docker Configuration

Kubernetes Deployment

Prerequisites

Deploy to Kubernetes

Scale Deployment

Update Deployment

AWS Deployment

Using CloudFormation

Using EC2 Directly

Using ECS

Production Best Practices

Security

Performance

Monitoring

Backup and Recovery

High Availability

Updates and Maintenance

Troubleshooting Deployment

Common Issues

Health Checks

Support