egullickson/motovaultpro
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

k8s prepwork

Browse Source
This commit is contained in:
Eric gullickson
2025-09-18 11:04:41 -05:00
parent 17d27f4b92
commit cb98336d5e
16 changed files with 555 additions and 434 deletions
Download Patch File Download Diff File Expand all files Collapse all files

629
K8S-REDESIGN.md
View File

@@ -2,20 +2,20 @@
## Overview
This document outlines the comprehensive redesign of MotoVaultPro's Docker Compose architecture to closely replicate a Kubernetes deployment pattern. The goal is to maintain all current functionality while preparing for seamless K8s migration and improving development experience.
This document outlines the aggressive redesign of MotoVaultPro's Docker Compose architecture to closely replicate a Kubernetes deployment pattern. **Breaking changes are acceptable** as this is a pre-production application. The goal is to completely replace the current architecture with a production-ready K8s-equivalent setup in 2-3 days, eliminating all development shortcuts and implementing true production constraints.
**SCOPE**: ETL services have been completely removed from the architecture. This migration covers the 11 remaining core services with a focus on security, observability, and K8s compatibility over backward compatibility.
## Current Architecture Analysis
### Existing Services (13 containers)
### Core Services for Migration (11 containers)
**MVP Platform Services (Microservices)**
- `mvp-platform-landing` - Marketing/landing page (nginx)
- `mvp-platform-tenants` - Multi-tenant management API (FastAPI)
- `mvp-platform-vehicles-api` - Vehicle data API (FastAPI)
- `mvp-platform-vehicles-etl` - Data processing pipeline (Python)
- `mvp-platform-vehicles-db` - Vehicle data storage (PostgreSQL)
- `mvp-platform-vehicles-redis` - Vehicle data cache (Redis)
- `mvp-platform-vehicles-mssql` - Monthly ETL source (SQL Server)
**Application Services (Modular Monolith)**
- `admin-backend` - Application API with feature capsules (Node.js)
@@ -25,66 +25,97 @@ This document outlines the comprehensive redesign of MotoVaultPro's Docker Compo
- `admin-minio` - Object storage (MinIO)
**Infrastructure**
- `nginx-proxy` - Load balancer and SSL termination
- `platform-postgres` - Platform services database
- `platform-redis` - Platform services cache
- `nginx-proxy` - **TO BE COMPLETELY REMOVED** (replaced by Traefik)
### Current Limitations
### Current Limitations (TO BE BROKEN)
1. **Single Network**: All services on default network
2. **Manual Routing**: nginx configuration requires manual updates
3. **Port Exposure**: Many services expose ports directly
4. **Configuration**: Environment variables scattered across services
5. **Service Discovery**: Hard-coded service names
6. **Observability**: Limited monitoring and debugging capabilities
1. **Single Network**: All services on default network - **BREAKING: Move to isolated networks**
2. **Manual Routing**: nginx configuration requires manual updates - **BREAKING: Complete removal**
3. **Excessive Port Exposure**: 10+ services expose ports directly - **BREAKING: Remove all except Traefik**
4. **Environment Variable Configuration**: 35+ env vars scattered across services - **BREAKING: Mandatory file-based config**
5. **Development Shortcuts**: Debug modes, open CORS, no authentication - **BREAKING: Production-only mode**
6. **No Resource Limits**: Services can consume unlimited resources - **BREAKING: Enforce limits on all services**
## Target Kubernetes-like Architecture
### Network Segmentation
### Network Segmentation (Aggressive Isolation)
```yaml
networks:
frontend:
driver: bridge
internal: false # Only for Traefik public access
labels:
- "com.motovaultpro.network=frontend"
- "com.motovaultpro.purpose=public-facing"
- "com.motovaultpro.purpose=public-traffic-only"
backend:
driver: bridge
internal: true
internal: true # Complete isolation from host
labels:
- "com.motovaultpro.network=backend"
- "com.motovaultpro.purpose=api-services"
database:
driver: bridge
internal: true
internal: true # Application data isolation
labels:
- "com.motovaultpro.network=database"
- "com.motovaultpro.purpose=data-layer"
- "com.motovaultpro.purpose=app-data-layer"
platform:
driver: bridge
internal: true
internal: true # Platform microservices isolation
labels:
- "com.motovaultpro.network=platform"
- "com.motovaultpro.purpose=microservices"
- "com.motovaultpro.purpose=platform-services"
```
### Service Placement Strategy
**BREAKING CHANGE**: No `egress` network. Services requiring external API access (Auth0, Google Maps, VPIC) will connect through the `backend` network with Traefik handling external routing. This forces all external communication through the ingress controller, matching Kubernetes egress gateway patterns.
| Network | Services | Purpose | K8s Equivalent |
### Service Placement Strategy (Aggressive Isolation)
| Service | Networks | Purpose | K8s Equivalent |
|---------|----------|---------|----------------|
| `frontend` | traefik, admin-frontend, mvp-platform-landing | Public-facing services | Public LoadBalancer services |
| `backend` | admin-backend, mvp-platform-tenants, mvp-platform-vehicles-api | API services | ClusterIP services |
| `database` | All PostgreSQL, Redis, MinIO | Data persistence | StatefulSets with PVCs |
| `platform` | Platform microservices communication | Internal service mesh | Service mesh networking |
| `traefik` | `frontend`, `backend` | **ONLY** public routing + API access | LoadBalancer + IngressController |
| `admin-frontend`, `mvp-platform-landing` | `frontend` | Public web applications | Ingress frontends |
| `admin-backend` | `backend`, `database`, `platform` | Application API with cross-service access | ClusterIP with multiple network attachment |
| `mvp-platform-tenants`, `mvp-platform-vehicles-api` | `backend`, `platform` | Platform APIs + data access | ClusterIP (platform namespace) |
| `admin-postgres`, `admin-redis`, `admin-minio` | `database` | Application data isolation | StatefulSets with PVCs |
| `platform-postgres`, `platform-redis`, `mvp-platform-vehicles-db`, `mvp-platform-vehicles-redis` | `platform` | Platform data isolation | StatefulSets with PVCs |
**BREAKING CHANGES**:
- **No external network access** for individual services
- **No host port exposure** except Traefik (80, 443, 8080)
- **Mandatory network isolation** - services cannot access unintended networks
- **No development bypasses** - all traffic through Traefik
**Service Communication Matrix (Restricted)**
```
# Internal service communication (via backend network)
admin-backend → mvp-platform-vehicles-api:8000 (authenticated API calls)
admin-backend → mvp-platform-tenants:8000 (authenticated API calls)
# Data layer access (isolated networks)
admin-backend → admin-postgres:5432, admin-redis:6379, admin-minio:9000
mvp-platform-vehicles-api → mvp-platform-vehicles-db:5432, mvp-platform-vehicles-redis:6379
mvp-platform-tenants → platform-postgres:5432, platform-redis:6379
# External integrations (BREAKING: via Traefik proxy only)
admin-backend → External APIs (Auth0, Google Maps, VPIC) via Traefik middleware
Platform services → External APIs via Traefik middleware (no direct access)
```
**BREAKING CHANGE**: All external API calls must be proxied through Traefik middleware. No direct external network access for any service.
## Traefik Configuration
### Core Traefik Setup
- New directories `config/traefik/` and `secrets/traefik/` will store production-bound configuration and certificates. These folders are justified as they mirror their eventual Kubernetes ConfigMap/Secret counterparts and replace the legacy nginx configuration.
```yaml
traefik:
image: traefik:v3.0
@@ -98,15 +129,15 @@ traefik:
- "8080:8080" # Dashboard
volumes:
- /var/run/docker.sock:/var/run/docker.sock:ro
- ./config/traefik:/config:ro
- ./certs:/certs:ro
configs:
- source: traefik-config
target: /etc/traefik/traefik.yml
- ./config/traefik/traefik.yml:/etc/traefik/traefik.yml:ro
- ./config/traefik/middleware.yml:/etc/traefik/middleware.yml:ro
- ./secrets/traefik/certs:/certs:ro
labels:
- "traefik.enable=true"
- "traefik.http.routers.dashboard.rule=Host(`traefik.motovaultpro.local`)"
- "traefik.http.routers.dashboard.tls=true"
- "traefik.http.routers.dashboard.middlewares=dashboard-allowlist@docker"
- "traefik.http.middlewares.dashboard-allowlist.ipwhitelist.sourcerange=10.0.0.0/8,172.16.0.0/12,192.168.0.0/16"
```
### Service Discovery Labels
@@ -197,6 +228,12 @@ http:
authResponseHeaders:
- "X-Auth-User"
- "X-Auth-Roles"
dashboard-allowlist:
ipWhiteList:
sourceRange:
- "10.0.0.0/8"
- "172.16.0.0/12"
- "192.168.0.0/16"
```
## Enhanced Health Checks
@@ -241,60 +278,101 @@ All services must expose:
## Configuration Management
### Docker Configs (K8s ConfigMaps equivalent)
### Configuration & Secret Management (Compose-compatible)
```yaml
configs:
traefik-config:
file: ./config/traefik/traefik.yml
traefik-middleware:
file: ./config/traefik/middleware.yml
app-config-production:
file: ./config/app/production.yml
platform-config:
file: ./config/platform/services.yml
- Application and platform settings will live in versioned files under `config/app/` and `config/platform/`, mounted read-only into the containers (`volumes:`). This mirrors ConfigMaps without relying on Docker Swarm-only `configs`.
- Secrets (Auth0, database, API keys) will be stored as individual files beneath `secrets/app/` and `secrets/platform/`, mounted as read-only volumes. At runtime the containers will read from `/run/secrets/*`, matching the eventual Kubernetes Secret mount pattern.
- Committed templates: `.example` files now reside in `config/app/production.yml.example`, `config/platform/production.yml.example`, and `secrets/**/.example` to document required keys while keeping live credentials out of Git. The real files stay untracked via `.gitignore`.
- Runtime loader: extend `backend/src/core/config/environment.ts` (and equivalent FastAPI settings) to hydrate configuration by reading `CONFIG_PATH` YAML and `SECRETS_DIR` file values before falling back to `process.env`. This ensures parity between Docker Compose mounts and future Kubernetes ConfigMap/Secret projections.
#### Configuration Migration Strategy
**Current Environment Variables (45 total) to File Mapping:**
**Application Secrets** (`secrets/app/`):
```
auth0-client-secret.txt # AUTH0_CLIENT_SECRET
postgres-password.txt # DB_PASSWORD
minio-access-key.txt # MINIO_ACCESS_KEY
minio-secret-key.txt # MINIO_SECRET_KEY
platform-vehicles-api-key.txt # PLATFORM_VEHICLES_API_KEY
google-maps-api-key.txt # GOOGLE_MAPS_API_KEY
```
### Docker Secrets (K8s Secrets equivalent)
```yaml
secrets:
auth0-client-secret:
file: ./secrets/auth0-client-secret.txt
database-passwords:
file: ./secrets/database-passwords.txt
platform-api-keys:
file: ./secrets/platform-api-keys.txt
ssl-certificates:
file: ./secrets/ssl-certs.txt
**Platform Secrets** (`secrets/platform/`):
```
platform-db-password.txt # PLATFORM_DB_PASSWORD
vehicles-db-password.txt # POSTGRES_PASSWORD (vehicles)
```
### Environment Configuration
**Network attachments for outbound-enabled services:**
```yaml
# config/app/production.yml
mvp-platform-vehicles-api:
networks:
- backend
- platform
- egress
mvp-platform-tenants:
networks:
- backend
- platform
- egress
```
**Application Configuration** (`config/app/production.yml`):
```yaml
server:
port: 3001
tenant_id: admin
database:
host: admin-postgres
port: 5432
name: motovaultpro
pool_size: 20
user: postgres
redis:
host: admin-redis
port: 6379
db: 0
minio:
endpoint: admin-minio
port: 9000
bucket: motovaultpro
auth0:
domain: ${AUTH0_DOMAIN}
audience: ${AUTH0_AUDIENCE}
domain: motovaultpro.us.auth0.com
audience: https://api.motovaultpro.com
platform:
vehicles_api:
url: http://mvp-platform-vehicles-api:8000
timeout: 30s
tenants_api:
url: http://mvp-platform-tenants:8000
timeout: 15s
vehicles_api_url: http://mvp-platform-vehicles-api:8000
tenants_api_url: http://mvp-platform-tenants:8000
external:
vpic_api_url: https://vpic.nhtsa.dot.gov/api/vehicles
```
**Compose Example:**
```yaml
admin-backend:
volumes:
- ./config/app/production.yml:/app/config/production.yml:ro
- ./secrets/app/auth0-client-secret.txt:/run/secrets/auth0-client-secret:ro
- ./secrets/app/postgres-password.txt:/run/secrets/postgres-password:ro
- ./secrets/app/minio-access-key.txt:/run/secrets/minio-access-key:ro
- ./secrets/app/minio-secret-key.txt:/run/secrets/minio-secret-key:ro
- ./secrets/app/platform-vehicles-api-key.txt:/run/secrets/platform-vehicles-api-key:ro
- ./secrets/app/google-maps-api-key.txt:/run/secrets/google-maps-api-key:ro
environment:
- NODE_ENV=production
- CONFIG_PATH=/app/config/production.yml
- SECRETS_DIR=/run/secrets
networks:
- backend
- database
- platform
- egress
```
## Resource Management
@@ -303,34 +381,23 @@ platform:
**Tier 1: Critical Services**
```yaml
deploy:
resources:
limits: { memory: 2G, cpus: '2.0' }
reservations: { memory: 1G, cpus: '1.0' }
restart_policy:
condition: on-failure
max_attempts: 3
admin-backend:
mem_limit: 2g
cpus: 2.0
```
**Tier 2: Supporting Services**
```yaml
deploy:
resources:
limits: { memory: 1G, cpus: '1.0' }
reservations: { memory: 512M, cpus: '0.5' }
restart_policy:
condition: on-failure
max_attempts: 3
admin-frontend:
mem_limit: 1g
cpus: 1.0
```
**Tier 3: Infrastructure Services**
```yaml
deploy:
resources:
limits: { memory: 512M, cpus: '0.5' }
reservations: { memory: 256M, cpus: '0.25' }
restart_policy:
condition: unless-stopped
traefik:
mem_limit: 512m
cpus: 0.5
```
### Service Tiers
@@ -339,126 +406,299 @@ deploy:
|------|----------|------------------|----------|
| 1 | admin-backend, mvp-platform-vehicles-api, admin-postgres | High | Critical |
| 2 | admin-frontend, mvp-platform-tenants, mvp-platform-landing | Medium | Important |
| 3 | traefik, redis services, etl services | Low | Supporting |
| 3 | traefik, redis services, storage services | Low | Supporting |
## Migration Implementation Plan
### Development Port Exposure Policy
### Phase 1: Infrastructure Foundation (Week 1)
**Exposed Ports for Development Debugging:**
```yaml
# Database Access (development debugging)
- 5432:5432 # admin-postgres (application DB access)
- 5433:5432 # mvp-platform-vehicles-db (platform DB access)
- 5434:5432 # platform-postgres (platform services DB access)
**Objectives:**
- Implement Traefik service
- Create network segmentation
- Establish basic routing
# Cache Access (development debugging)
- 6379:6379 # admin-redis
- 6380:6379 # mvp-platform-vehicles-redis
- 6381:6379 # platform-redis
# Storage Access (development/admin)
- 9000:9000 # admin-minio API
- 9001:9001 # admin-minio console
# Traefik Dashboard (development monitoring)
- 8080:8080 # traefik dashboard
```
**Internal-Only Services (no port exposure):**
- All HTTP application services (routed through Traefik)
- Platform APIs (accessible via application backend only)
**Mobile Testing Considerations:**
- Self-signed certificates require device-specific trust configuration
- Development URLs must be accessible from mobile devices on same network
- Certificate CN must match both `motovaultpro.com` and `admin.motovaultpro.com`
## Migration Implementation Plan (Aggressive Approach)
### **BREAKING CHANGE STRATEGY**: Complete Architecture Replacement (2-3 Days)
**Objective**: Replace entire Docker Compose architecture with K8s-equivalent setup in a single migration event. No backward compatibility, no gradual transition, no service uptime requirements.
### **Day 1: Complete Infrastructure Replacement**
**Breaking Changes Implemented:**
1. **Remove nginx-proxy completely** - no parallel operation
2. **Implement Traefik with full production configuration**
3. **Break all current networking** - implement 4-network isolation from scratch
4. **Remove ALL development port exposure** (10+ ports → 3 ports)
5. **Break environment variable patterns** - implement mandatory file-based configuration
**Tasks:**
1. Create new network topology
2. Add Traefik service with basic configuration
3. Migrate nginx routing to Traefik labels
4. Test SSL certificate handling
5. Verify all existing functionality
```bash
# 1. Backup current state
cp docker-compose.yml docker-compose.old.yml
docker compose down
**Success Criteria:**
- All services accessible via original URLs
- SSL certificates working
- Health checks functional
- No performance degradation
# 2. Create configuration structure
mkdir -p config/app config/platform secrets/app secrets/platform
### Phase 2: Service Discovery & Labels (Week 2)
# 3. Generate production-ready certificates
make generate-certs # Multi-domain with mobile compatibility
**Objectives:**
- Convert all services to label-based discovery
- Implement middleware for security
- Add service health monitoring
# 4. Implement new docker-compose.yml with:
# - 4 isolated networks
# - Traefik service with full middleware
# - No port exposure except Traefik (80, 443, 8080)
# - File-based configuration for all services
# - Resource limits on all services
# 5. Update all service configurations to use file-based config
# - Remove all environment variables from compose
# - Implement CONFIG_PATH and SECRETS_DIR loaders
```
**Expected Failures**: Services will fail to start until configuration files are properly implemented.
### **Day 2: Service Reconfiguration & Authentication**
**Breaking Changes Implemented:**
1. **Mandatory service-to-service authentication** - remove all debug/open access
2. **Implement standardized health endpoints** - break existing health check patterns
3. **Enforce resource limits** - services may fail if exceeding limits
4. **Remove CORS development shortcuts** - production-only security
**Tasks:**
1. Convert each service to Traefik labels
2. Implement security middleware
3. Add CORS and authentication middleware
4. Test service discovery and failover
5. Implement Traefik dashboard access
```bash
# 1. Implement /health, /health/ready, /health/live on all HTTP services
# 2. Update Dockerfiles and service code for new health endpoints
# 3. Configure Traefik labels for all services
# 4. Implement service authentication:
# - API keys for platform service access
# - Remove debug modes and localhost CORS
# - Implement production security headers
# 5. Add resource limits to all services
# 6. Test new architecture end-to-end
```
**Success Criteria:**
- All services discovered automatically
- Security middleware working
- Dashboard accessible and functional
- Mobile and desktop testing passes
**Expected Issues**: Authentication failures, CORS errors, resource limit violations.
### Phase 3: Configuration Management (Week 3)
**Objectives:**
- Implement Docker configs and secrets
- Standardize environment configuration
- Add monitoring and observability
### **Day 3: Validation & Documentation Update**
**Tasks:**
1. Move configuration to Docker configs
2. Implement secrets management
3. Standardize health check endpoints
4. Add service metrics collection
5. Implement log aggregation
1. **Complete testing** of new architecture
2. **Update all documentation** to reflect new constraints
3. **Update Makefile** with breaking changes to commands
4. **Validate mobile access** with new certificate and routing
5. **Performance validation** (baseline not required - new architecture is target)
**Success Criteria:**
- No hardcoded secrets in compose files
- Centralized configuration management
- Enhanced monitoring capabilities
- Improved debugging experience
### **BREAKING CHANGES SUMMARY**
### Phase 4: Optimization & Documentation (Week 4)
#### **Network Access**
- **OLD**: All services on default network with host access
- **NEW**: 4 isolated networks, no host access except Traefik
**Objectives:**
- Optimize resource allocation
- Update development workflow
- Complete documentation
#### **Port Exposure**
- **OLD**: 10+ ports exposed (databases, APIs, storage)
- **NEW**: Only 3 ports (80, 443, 8080) - everything through Traefik
**Tasks:**
1. Implement resource limits and reservations
2. Update Makefile with new commands
3. Create troubleshooting documentation
4. Performance testing and optimization
5. Final validation of all features
#### **Configuration**
- **OLD**: 35+ environment variables scattered across services
- **NEW**: Mandatory file-based configuration with no env fallbacks
**Success Criteria:**
- Optimized resource usage
- Updated development workflow
- Complete documentation
- All tests passing
#### **Development Access**
- **OLD**: Direct database/service access via exposed ports
- **NEW**: Access only via `docker exec` or Traefik routing
## Development Workflow Enhancements
#### **Security**
- **OLD**: Debug modes, open CORS, no authentication
- **NEW**: Production security only, mandatory authentication
### New Makefile Commands
#### **Resource Management**
- **OLD**: Unlimited resource consumption
- **NEW**: Enforced limits on all services
### **Risk Mitigation**
1. **Document current working state** before migration (Day 0)
2. **Keep docker-compose.old.yml** for reference
3. **Backup all volumes** before starting
4. **Expect multiple restart cycles** during configuration
5. **Plan for debugging time** - new constraints will reveal issues
### **Success Criteria (Non-Negotiable)**
- ✅ All 11 services operational through Traefik only
- ✅ Zero host port exposure except Traefik
- ✅ All configuration file-based
- ✅ Service-to-service authentication working
- ✅ Mobile and desktop HTTPS access functional
- ✅ Resource limits enforced and services stable
## Development Workflow Enhancements (BREAKING CHANGES)
### Updated Makefile Commands (BREAKING CHANGES)
**BREAKING CHANGE**: All database and service direct access removed. New K8s-equivalent workflow only.
**Core Commands (Updated for New Architecture):**
```makefile
SHELL := /bin/bash
# Traefik specific commands
traefik-dashboard:
@echo "Opening Traefik dashboard..."
@open https://traefik.motovaultpro.local:8080
@echo "Traefik dashboard: http://localhost:8080"
@echo "Add to /etc/hosts: 127.0.0.1 traefik.motovaultpro.local"
traefik-logs:
@docker compose logs -f traefik
service-discovery:
@echo "Discovered services:"
@docker compose exec traefik traefik api --url=http://localhost:8080/api/rawdata
@echo "Discovered services and routes:"
@docker compose exec traefik curl -sf http://localhost:8080/api/rawdata | jq '.http.services, .http.routers' 2>/dev/null || docker compose exec traefik curl -sf http://localhost:8080/api/rawdata
network-inspect:
@echo "Network topology:"
@docker network ls --filter name=motovaultpro
@docker network inspect motovaultpro_frontend motovaultpro_backend motovaultpro_database motovaultpro_platform
@docker network inspect motovaultpro_frontend motovaultpro_backend motovaultpro_database motovaultpro_platform motovaultpro_egress 2>/dev/null | jq '.[].Name, .[].Containers' || echo "Networks not yet created"
health-check-all:
@echo "Checking health of all services..."
@docker compose ps --format "table {{.Service}}\t{{.Status}}\t{{.Health}}"
# Enhanced existing commands
# Mobile testing support
mobile-setup:
@echo "Mobile Testing Setup:"
@echo "1. Connect mobile device to same network as development machine"
@echo "2. Find development machine IP: $$(hostname -I | awk '{print $$1}')"
@echo "3. Add to mobile device hosts file (if rooted) or use IP directly:"
@echo " $$(hostname -I | awk '{print $$1}') motovaultpro.com"
@echo " $$(hostname -I | awk '{print $$1}') admin.motovaultpro.com"
@echo "4. Install certificate from: https://$$(hostname -I | awk '{print $$1}')/certs/motovaultpro.com.crt"
@echo "5. Trust certificate in device settings"
# Development database access
db-admin:
@echo "Database Access:"
@echo "Application DB: postgresql://postgres:localdev123@localhost:5432/motovaultpro"
@echo "Platform DB: postgresql://platform_user:platform123@localhost:5434/platform"
@echo "Vehicles DB: postgresql://mvp_platform_user:platform123@localhost:5433/vehicles"
db-shell-app:
@docker compose exec admin-postgres psql -U postgres -d motovaultpro
db-shell-platform:
@docker compose exec platform-postgres psql -U platform_user -d platform
db-shell-vehicles:
@docker compose exec mvp-platform-vehicles-db psql -U mvp_platform_user -d vehicles
# Enhanced existing commands (preserve ETL removal)
logs:
@echo "Available log targets: all, traefik, backend, frontend, platform"
@echo "Available log targets: all, traefik, backend, frontend, platform, vehicles-api, tenants"
@docker compose logs -f $(filter-out $@,$(MAKECMDGOALS))
# Remove ETL commands
# etl-load-manual, etl-load-clear, etl-validate-json, etl-shell - REMOVED (out of scope)
%:
@: # This catches the log target argument
```
### Enhanced Development Features
**Updated Core Commands:**
```makefile
setup:
@echo "Setting up MotoVaultPro K8s-ready development environment..."
@echo "1. Checking configuration files..."
@if [ ! -d config ]; then echo "Creating config directory structure..."; mkdir -p config/app config/platform secrets/app secrets/platform; fi
@echo "2. Checking SSL certificates..."
@if [ ! -f certs/motovaultpro.com.crt ]; then echo "Generating multi-domain SSL certificate..."; $(MAKE) generate-certs; fi
@echo "3. Building and starting all containers..."
@docker compose up -d --build --remove-orphans
@echo "4. Running database migrations..."
@sleep 15 # Wait for databases to be ready
@docker compose exec admin-backend node dist/_system/migrations/run-all.js
@echo ""
@echo "✅ K8s-ready setup complete!"
@echo "Access application at: https://admin.motovaultpro.com"
@echo "Access platform landing at: https://motovaultpro.com"
@echo "Traefik dashboard: http://localhost:8080"
@echo "Mobile setup: make mobile-setup"
generate-certs:
@echo "Generating multi-domain SSL certificate for mobile compatibility..."
@mkdir -p certs
@openssl req -x509 -nodes -days 365 -newkey rsa:2048 \
-keyout certs/motovaultpro.com.key \
-out certs/motovaultpro.com.crt \
-config <(echo '[dn]'; echo 'CN=motovaultpro.com'; echo '[req]'; echo 'distinguished_name = dn'; echo '[SAN]'; echo 'subjectAltName=DNS:motovaultpro.com,DNS:admin.motovaultpro.com,DNS:*.motovaultpro.com,IP:127.0.0.1') \
-extensions SAN
@echo "Certificate generated with SAN for mobile compatibility"
# New K8s-equivalent access patterns
db-access:
@echo "🚫 BREAKING CHANGE: No direct port access"
@echo "Database access via container exec only:"
@echo " Application DB: make db-shell-app"
@echo " Platform DB: make db-shell-platform"
@echo " Vehicles DB: make db-shell-vehicles"
# Service inspection (K8s equivalent)
service-status:
@echo "Service health status:"
@docker compose ps --format "table {{.Service}}\\t{{.Status}}\\t{{.Health}}"
traefik-dashboard:
@echo "Traefik Dashboard: http://localhost:8080"
# Mobile testing (updated for new architecture)
mobile-setup:
@echo "📱 Mobile Testing Setup (New Architecture):"
@echo "1. Connect mobile device to same network"
@echo "2. Development machine IP: $$(hostname -I | awk '{print $$1}')"
@echo "3. Add DNS: $$(hostname -I | awk '{print $$1}') motovaultpro.com admin.motovaultpro.com"
@echo "4. Trust certificate and access: https://admin.motovaultpro.com"
# REMOVED COMMANDS (Breaking changes):
# ❌ All direct port access commands
# ❌ ETL commands (out of scope)
# ❌ Development shortcuts
```
### **BREAKING CHANGES TO DEVELOPMENT WORKFLOW**
#### **Database Access**
- **OLD**: `psql -h localhost -p 5432` (direct connection)
- **NEW**: `make db-shell-app` (container exec only)
#### **Service Debugging**
- **OLD**: `curl http://localhost:8000/health` (direct port)
- **NEW**: `curl https://admin.motovaultpro.com/api/platform/vehicles/health` (via Traefik)
#### **Storage Access**
- **OLD**: MinIO console at `http://localhost:9001`
- **NEW**: Access via Traefik routing only
### Enhanced Development Features (Updated)
**Service Discovery Dashboard**
- Real-time service status
@@ -473,10 +713,12 @@ logs:
- Performance metrics
**Testing Enhancements**
- Automated health checks
- Service integration testing
- Load balancing validation
- SSL certificate verification
- Automated health checks across all services
- Service integration testing with network isolation
- Load balancing validation through Traefik
- SSL certificate verification for desktop and mobile
- Mobile device testing workflow validation
- Cross-network service communication testing
## Observability & Monitoring
@@ -507,6 +749,32 @@ metrics:
### Health Monitoring
**Service Health Dashboard**
- Real-time service status via Traefik dashboard
- Historical health trends (Phase 4 enhancement)
- Network connectivity validation
- Mobile accessibility monitoring
**Critical Monitoring Points:**
1. **Service Discovery**: All services registered with Traefik
2. **Network Isolation**: Services only accessible via designated networks
3. **SSL Certificate Status**: Valid certificates for all domains
4. **Mobile Compatibility**: Certificate trust and network accessibility
5. **Database Connectivity**: Cross-network database access patterns
6. **Platform API Authentication**: Service-to-service authentication working
**Development Health Checks:**
```bash
# Quick health validation
make health-check-all
make service-discovery
make network-inspect
# Mobile testing validation
make mobile-setup
curl -k https://admin.motovaultpro.com/health # From mobile device IP
```
**Service Health Dashboard**
- Real-time service status
- Historical health trends
@@ -601,11 +869,13 @@ metrics:
### Backup Strategy
- Backup current docker-compose.yml
- Backup nginx configuration
- Export service configurations
- Document current network topology
- Save working environment variables
**Critical Data Backup:**
- Backup platform services PostgreSQL database:
```bash
docker compose exec platform-postgres pg_dump -U platform_user platform > platform_backup_$(date +%Y%m%d_%H%M%S).sql
```
**Note:** All other services are stateless or use development data that can be recreated. Application database, Redis, and MinIO contain only development data.
## Success Metrics
@@ -650,6 +920,23 @@ metrics:
## Conclusion
This comprehensive redesign transforms the Docker Compose architecture to closely mirror Kubernetes deployment patterns while maintaining all existing functionality and improving the development experience. The phased migration approach ensures minimal disruption while delivering immediate benefits in observability, security, and maintainability.
This aggressive redesign completely replaces the Docker Compose architecture with a production-ready K8s-equivalent setup in 2-3 days. **Breaking changes are the strategy** - eliminating all development shortcuts and implementing true production constraints from day one.
The new architecture provides a solid foundation for future Kubernetes migration while enhancing current development workflows with modern service discovery, monitoring, and configuration management practices.
### **Key Transformation**
- **11 services** migrated from single-network to 4-network isolation
- **10+ exposed ports** reduced to 3 (Traefik only)
- **35+ environment variables** replaced with mandatory file-based configuration
- **All development bypasses removed** - production security enforced
- **Direct service access eliminated** - all traffic through Traefik
### **Benefits of Aggressive Approach**
1. **Faster Implementation**: 2-3 days vs 4 weeks of gradual migration
2. **Authentic K8s Simulation**: True production constraints from start
3. **No Legacy Debt**: Clean architecture without compatibility layers
4. **Better Security**: Production-only mode eliminates development vulnerabilities
5. **Simplified Testing**: Single target architecture instead of multiple transition states
### **Post-Migration State**
The new architecture provides an exact Docker Compose equivalent of Kubernetes deployment patterns. All services operate under production constraints with proper isolation, authentication, and resource management. This setup can be directly translated to Kubernetes manifests with minimal changes.
**Development teams gain production-like experience while maintaining local development efficiency through container-based workflows and Traefik-based service discovery.**