HomeAudit/migration_scripts/scripts/setup_docker_swarm.sh

#!/bin/bash
# Setup Docker Swarm Cluster
# This script initializes Docker Swarm across all hosts

set -euo pipefail

echo "🐳 Setting up Docker Swarm cluster..."

# Define hosts and their roles
MANAGER_HOST="omv800"
MANAGER_IP="192.168.50.229"
WORKER_HOSTS=("fedora" "surface" "jonathan-2518f5u" "audrey")
WORKER_IPS=("192.168.50.225" "192.168.50.254" "192.168.50.181" "192.168.50.145")

# Colors for output
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m' # No Color

# Function to print colored output
print_status() {
    echo -e "${GREEN}[INFO]${NC} $1"
}

print_warning() {
    echo -e "${YELLOW}[WARNING]${NC} $1"
}

print_error() {
    echo -e "${RED}[ERROR]${NC} $1"
}

# Function to check if Docker is installed
check_docker() {
    local host=$1
    print_status "Checking Docker installation on $host..."
    
    if ssh -o ConnectTimeout=10 "$host" "docker --version" > /dev/null 2>&1; then
        print_status "Docker is installed on $host"
        return 0
    else
        print_error "Docker is not installed on $host"
        return 1
    fi
}

# Function to check if host is already in swarm
check_swarm_status() {
    local host=$1
    if ssh -o ConnectTimeout=10 "$host" "docker info --format '{{.Swarm.LocalNodeState}}'" 2>/dev/null | grep -q "active"; then
        print_warning "$host is already part of a swarm"
        return 0
    else
        print_status "$host is not in swarm mode"
        return 1
    fi
}

# Function to leave swarm if already joined
leave_swarm() {
    local host=$1
    print_status "Leaving existing swarm on $host..."
    ssh -o ConnectTimeout=10 "$host" "docker swarm leave --force" 2>/dev/null || true
    sleep 5
}

# 1. Check Docker installation on all hosts
print_status "Step 1: Checking Docker installation..."
for host in "$MANAGER_HOST" "${WORKER_HOSTS[@]}"; do
    if ! check_docker "$host"; then
        print_error "Please install Docker on $host before proceeding"
        exit 1
    fi
done

# 2. Initialize swarm on manager
print_status "Step 2: Initializing swarm on manager ($MANAGER_HOST)..."
if check_swarm_status "$MANAGER_HOST"; then
    leave_swarm "$MANAGER_HOST"
fi

ssh "$MANAGER_HOST" "docker swarm init --advertise-addr $MANAGER_IP --listen-addr $MANAGER_IP"

# Get join token for workers
print_status "Getting join token for workers..."
JOIN_TOKEN=$(ssh "$MANAGER_HOST" "docker swarm join-token -q worker")
MANAGER_TOKEN=$(ssh "$MANAGER_HOST" "docker swarm join-token -q manager")

print_status "Worker token: $JOIN_TOKEN"
print_status "Manager token: $MANAGER_TOKEN"

# 3. Join workers to swarm
print_status "Step 3: Joining workers to swarm..."
for i in "${!WORKER_HOSTS[@]}"; do
    host="${WORKER_HOSTS[$i]}"
    ip="${WORKER_IPS[$i]}"
    
    print_status "Joining $host ($ip) to swarm..."
    
    if check_swarm_status "$host"; then
        leave_swarm "$host"
    fi
    
    if ssh -o ConnectTimeout=10 "$host" "docker swarm join --token $JOIN_TOKEN $MANAGER_IP:2377 --advertise-addr $ip --listen-addr $ip"; then
        print_status "Successfully joined $host to swarm"
    else
        print_error "Failed to join $host to swarm"
        exit 1
    fi
done

# 4. Verify swarm status
print_status "Step 4: Verifying swarm status..."
sleep 10

print_status "Swarm nodes:"
ssh "$MANAGER_HOST" "docker node ls"

# 5. Create overlay networks
print_status "Step 5: Creating overlay networks..."

NETWORKS=(
    "traefik-public"
    "monitoring"
    "databases"
    "applications"
    "iot-network"
    "backup-network"
)

for network in "${NETWORKS[@]}"; do
    print_status "Creating network: $network"
    if ssh "$MANAGER_HOST" "docker network create --driver overlay --attachable $network" 2>/dev/null; then
        print_status "Created network: $network"
    else
        print_warning "Network $network may already exist"
    fi
done

# 6. Setup swarm manager backup
print_status "Step 6: Setting up manager backup..."
print_status "Promoting surface as backup manager..."

if ssh "surface" "docker swarm join --token $MANAGER_TOKEN $MANAGER_IP:2377 --advertise-addr 192.168.50.254 --listen-addr 192.168.50.254"; then
    print_status "Successfully promoted surface as backup manager"
else
    print_warning "Failed to promote surface as backup manager"
fi

# 7. Configure swarm settings
print_status "Step 7: Configuring swarm settings..."

# Set up auto-lock for security
ssh "$MANAGER_HOST" "docker swarm update --autolock=true"

# Configure logging
ssh "$MANAGER_HOST" "docker swarm update --log-driver=json-file --log-opt max-size=10m --log-opt max-file=3"

# 8. Create swarm configuration file
print_status "Step 8: Creating swarm configuration..."
cat > "/opt/migration/configs/swarm-config.yml" << EOF
# Docker Swarm Configuration
# Generated: $(date)

swarm:
  manager:
    primary: $MANAGER_HOST
    backup: surface
    ip: $MANAGER_IP
  
  workers:
$(for i in "${!WORKER_HOSTS[@]}"; do echo "    - host: ${WORKER_HOSTS[$i]}"; echo "      ip: ${WORKER_IPS[$i]}"; done)
  
  networks:
$(for network in "${NETWORKS[@]}"; do echo "    - $network"; done)
  
  tokens:
    worker: $JOIN_TOKEN
    manager: $MANAGER_TOKEN
  
  settings:
    autolock: true
    log_driver: json-file
    log_opts:
      max_size: 10m
      max_file: 3
EOF

# 9. Test swarm connectivity
print_status "Step 9: Testing swarm connectivity..."

# Test service deployment
print_status "Testing service deployment..."
ssh "$MANAGER_HOST" "docker service create --name test-service --replicas 2 --network traefik-public nginx:alpine"

sleep 10

# Check service status
print_status "Service status:"
ssh "$MANAGER_HOST" "docker service ls"
ssh "$MANAGER_HOST" "docker service ps test-service"

# Clean up test service
print_status "Cleaning up test service..."
ssh "$MANAGER_HOST" "docker service rm test-service"

# 10. Create health check script
print_status "Step 10: Creating health check script..."
cat > "/opt/migration/scripts/check_swarm_health.sh" << 'EOF'
#!/bin/bash
# Check Docker Swarm Health

set -euo pipefail

MANAGER_HOST="omv800"

echo "🏥 Checking Docker Swarm health..."

# Check node status
echo "📋 Node status:"
ssh "$MANAGER_HOST" "docker node ls"

# Check network status
echo "🌐 Network status:"
ssh "$MANAGER_HOST" "docker network ls --filter driver=overlay"

# Check service status
echo "🔧 Service status:"
ssh "$MANAGER_HOST" "docker service ls"

# Check swarm info
echo "ℹ️ Swarm info:"
ssh "$MANAGER_HOST" "docker info --format '{{.Swarm.LocalNodeState}}'"

echo "✅ Swarm health check completed"
EOF

chmod +x "/opt/migration/scripts/check_swarm_health.sh"

# 11. Final verification
print_status "Step 11: Final verification..."

print_status "Swarm nodes:"
ssh "$MANAGER_HOST" "docker node ls"

print_status "Overlay networks:"
ssh "$MANAGER_HOST" "docker network ls --filter driver=overlay"

print_status "Swarm info:"
ssh "$MANAGER_HOST" "docker info --format '{{.Swarm.LocalNodeState}}'"

# 12. Create summary
print_status "Step 12: Creating setup summary..."
cat > "/opt/migration/setup_summary.txt" << EOF
Docker Swarm Setup Summary
Generated: $(date)

Manager Node:
  Host: $MANAGER_HOST
  IP: $MANAGER_IP
  Status: Active

Backup Manager:
  Host: surface
  IP: 192.168.50.254
  Status: Active

Worker Nodes:
$(for i in "${!WORKER_HOSTS[@]}"; do echo "  - ${WORKER_HOSTS[$i]}: ${WORKER_IPS[$i]}"; done)

Networks Created:
$(for network in "${NETWORKS[@]}"; do echo "  - $network"; done)

Configuration Files:
  - /opt/migration/configs/swarm-config.yml
  - /opt/migration/scripts/check_swarm_health.sh

Next Steps:
1. Deploy Traefik reverse proxy
2. Setup monitoring stack
3. Begin service migration
EOF

print_status "✅ Docker Swarm setup completed successfully!"
print_status "📋 Setup summary saved to: /opt/migration/setup_summary.txt"
print_status "🔧 Health check script: /opt/migration/scripts/check_swarm_health.sh"

echo ""
print_status "Next steps:"
echo "  1. Deploy Traefik: ./scripts/deploy_traefik.sh"
echo "  2. Setup monitoring: ./scripts/setup_monitoring.sh"
echo "  3. Begin migration: ./scripts/start_migration.sh"