Qelos Deployment Guide

This guide explains how to deploy Qelos using Helm charts, covering both local development and production deployments.

Helm Chart Structure

The Qelos Helm chart is organized as follows:

helm/qelos/
├── Chart.yaml
├── values.yaml         # Default configuration values
├── values-env.yaml    # Environment-specific values
└── templates/
    ├── microservices.yaml
    └── _microservice.tpl

Configuration

Global Settings

The Helm chart provides several global configuration options in values.yaml:

global:
  environment: development
  internalSecret: "your-internal-secret"
  redis:
    host: redis://redis-service
    port: 6379
  mongodb:
    deployInCluster: true    # Set to false for external MongoDB
    external:
      host: ""
      port: 27017
      database: qelos

MongoDB Configuration

You can choose between deploying MongoDB within the cluster or using an external instance:

1. In-Cluster MongoDB (default):

global:
  mongodb:
    deployInCluster: true
    internal:
      host: mongodb-service
      port: 27017
      database: qelos

2. External MongoDB:

global:
  mongodb:
    deployInCluster: false
    external:
      host: "your-mongodb-host"
      port: 27017
      database: qelos

Service Configuration

Each microservice can be configured independently:

serviceName:
  image:
    repository: registry.gitlab.com/qelos/qelos/service-name
    tag: latest
  host: service-name-service
  port: 9000
  environment:
    PORT: 9000
    MONGO_URI: mongodb://mongodb-service:27017/qelos
    REDIS_URL: redis://redis-service
    SECRETS_TOKEN: ${SERVICE_SECRET}

Deployment Process

1. Prerequisites

• Kubernetes cluster (local or remote)
• Helm 3.x installed
• Access to container registry
• Environment variables file (.env)

2. Generate Helm Values

Before deployment, generate the environment-specific values:

node --env-file .env tools/ingest-helm-values/index.js

This creates values-env.yaml with environment-specific configurations.

3. Deploy

Deploy or upgrade Qelos using Helm:

helm upgrade --install qelos -f ./helm/qelos/values-env.yaml ./helm/qelos

Common deployment options:

# Dry run to verify configuration
helm upgrade --install --dry-run qelos -f ./helm/qelos/values-env.yaml ./helm/qelos

# Deploy to specific namespace
helm upgrade --install -n your-namespace qelos -f ./helm/qelos/values-env.yaml ./helm/qelos

# Set specific values during deployment
helm upgrade --install qelos -f ./helm/qelos/values-env.yaml \
  --set global.environment=production \
  --set auth.environment.SECRETS_TOKEN=your-secret \
  ./helm/qelos

4. Verify Deployment

# Check pod status
kubectl get pods

# View service logs
kubectl logs -f deployment/gateway-deployment

# Port forward gateway service (for local access)
kubectl port-forward svc/gateway-service 3000:80

Production Considerations

1. Security:

   • Never commit secrets to version control
   • Use Kubernetes secrets for sensitive data
   • Configure proper network policies

2. Resource Management:

   • Adjust resource requests/limits based on workload
   • Configure horizontal pod autoscaling
   • Monitor resource usage

3. High Availability:

   • Deploy across multiple availability zones
   • Configure proper replica counts
   • Use persistent storage for stateful services

4. Monitoring:

   • Set up monitoring and alerting
   • Configure proper logging
   • Implement health checks

Troubleshooting

Common issues and solutions:

1. Pod Startup Issues:

   kubectl describe pod <pod-name>
   kubectl logs <pod-name>

2. Service Discovery Issues:

   kubectl get svc
   kubectl describe svc <service-name>

3. Configuration Issues:

   • Verify values in values-env.yaml
   • Check environment variables
   • Validate service connectivity