#!/bin/bash set -e set -o pipefail # Get WC_ROOT (where this script and templates live) WC_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)" export WC_ROOT # ============================================================================= # HELPER FUNCTIONS (used by all phases) # ============================================================================= # Colors for output RED='\033[0;31m' GREEN='\033[0;32m' YELLOW='\033[1;33m' BLUE='\033[0;34m' NC='\033[0m' # No Color # Helper functions print_header() { echo -e "\n${BLUE}=== $1 ===${NC}\n" } print_info() { echo -e "${BLUE}INFO:${NC} $1" } print_warning() { echo -e "${YELLOW}WARNING:${NC} $1" } print_success() { echo -e "${GREEN}SUCCESS:${NC} $1" } print_error() { echo -e "${RED}ERROR:${NC} $1" } # Function to prompt for input with default value prompt_with_default() { local prompt="$1" local default="$2" local current_value="$3" local result if [ -n "${current_value}" ] && [ "${current_value}" != "null" ]; then printf "%s [current: %s]: " "${prompt}" "${current_value}" >&2 read -r result if [ -z "${result}" ]; then result="${current_value}" fi elif [ -n "${default}" ]; then printf "%s [default: %s]: " "${prompt}" "${default}" >&2 read -r result if [ -z "${result}" ]; then result="${default}" fi else printf "%s: " "${prompt}" >&2 read -r result while [ -z "${result}" ]; do printf "This value is required. Please enter a value: " >&2 read -r result done fi echo "${result}" } # Function to get current config value safely get_current_config() { local key="$1" if [ -f "${WC_HOME}/config.yaml" ]; then set +e result=$(wild-config "${key}" 2>/dev/null) set -e echo "${result}" else echo "" fi } # Function to get current secret value safely get_current_secret() { local key="$1" if [ -f "${WC_HOME}/secrets.yaml" ]; then set +e result=$(wild-secret "${key}" 2>/dev/null) set -e echo "${result}" else echo "" fi } UPDATE=false # Phase tracking variables SKIP_INSTALLER=false SKIP_HARDWARE=false SKIP_CONFIGS=false SKIP_INSTALL=false # Parse arguments while [[ $# -gt 0 ]]; do case $1 in --update) UPDATE=true shift ;; --skip-installer) SKIP_INSTALLER=true shift ;; --skip-hardware) SKIP_HARDWARE=true shift ;; --skip-configs) SKIP_CONFIGS=true shift ;; --skip-install) SKIP_INSTALL=true shift ;; --skip-all-phases) SKIP_INSTALLER=true SKIP_HARDWARE=true SKIP_CONFIGS=true SKIP_INSTALL=true shift ;; -h|--help) echo "Usage: $0 [--update] [phase-options]" echo "" echo "Initialize and set up a complete Wild-Cloud cluster deployment." echo "" echo "Cloud Options:" echo " --update Update existing cloud files (overwrite)" echo "" echo "Phase Control Options:" echo " --skip-installer Skip Phase 1 (Installer image generation)" echo " --skip-hardware Skip Phase 2 (Node hardware detection)" echo " --skip-configs Skip Phase 3 (Machine config generation)" echo " --skip-install Skip Phase 4 (Cluster services installation)" echo " --skip-all-phases Skip all phases (cloud setup only)" echo "" echo "General Options:" echo " -h, --help Show this help message" echo "" echo "Phases:" echo " 1. Installer image - Generate custom Talos installer URLs" echo " 2. Hardware detection - Discover node interfaces and disks" echo " 3. Machine configs - Generate Talos machine configurations" echo " 4. Cluster services - Install MetalLB, Traefik, cert-manager, etc." echo "" echo "Configuration is done automatically when needed by each phase." echo "" echo "By default, this script will only run in an empty directory." echo "Use --update to overwrite existing cloud files while preserving other files." exit 0 ;; -*) echo "Unknown option $1" echo "Usage: $0 [--update] [phase-options]" echo "Use --help for full usage information" exit 1 ;; *) echo "Unexpected argument: $1" echo "Usage: $0 [--update] [phase-options]" echo "Use --help for full usage information" exit 1 ;; esac done # Set up cloud directory (WC_HOME is where user's cloud will be) WC_HOME="$(pwd)" export WC_HOME # Template directory (in WC_ROOT, never written to) TEMPLATE_DIR="${WC_ROOT}/setup/home-scaffold" if [ ! -d "${TEMPLATE_DIR}" ]; then echo "Error: Template directory not found at ${TEMPLATE_DIR}" exit 1 fi # Check if cloud already exists if [ -d ".wildcloud" ]; then echo "Wild-Cloud already exists in this directory." echo "" read -p "Do you want to update cloud files? (y/N): " -n 1 -r echo if [[ $REPLY =~ ^[Yy]$ ]]; then UPDATE=true echo "Updating cloud files..." else echo "Skipping cloud update." echo "" fi else # Check if current directory is empty for new cloud if [ "${UPDATE}" = false ]; then # Check if directory has any files (including hidden files, excluding . and ..) if [ -n "$(find . -maxdepth 1 -name ".*" -o -name "*" | grep -v "^\.$" | head -1)" ]; then echo "Error: Current directory is not empty" echo "Use --update flag to overwrite existing cloud files while preserving other files" exit 1 fi fi echo "Initializing Wild-Cloud in $(pwd)" UPDATE=false fi # Initialize cloud files if needed if [ ! -d ".wildcloud" ] || [ "${UPDATE}" = true ]; then if [ "${UPDATE}" = true ]; then echo "Updating cloud files (preserving existing custom files)" else echo "Creating cloud files" fi # Function to copy files and directories copy_cloud_files() { local src_dir="$1" local dest_dir="$2" # Create destination directory if it doesn't exist mkdir -p "${dest_dir}" # Copy directory structure find "${src_dir}" -type d | while read -r src_subdir; do rel_path="${src_subdir#${src_dir}}" rel_path="${rel_path#/}" # Remove leading slash if present if [ -n "${rel_path}" ]; then mkdir -p "${dest_dir}/${rel_path}" fi done # Copy files find "${src_dir}" -type f | while read -r src_file; do rel_path="${src_file#${src_dir}}" rel_path="${rel_path#/}" # Remove leading slash if present dest_file="${dest_dir}/${rel_path}" # Ensure destination directory exists dest_file_dir=$(dirname "${dest_file}") mkdir -p "${dest_file_dir}" if [ "${UPDATE}" = true ] && [ -f "${dest_file}" ]; then echo "Updating: ${rel_path}" else echo "Creating: ${rel_path}" fi cp "${src_file}" "${dest_file}" done } # Copy cloud files to current directory copy_cloud_files "${TEMPLATE_DIR}" "." echo "" echo "Wild-Cloud initialized successfully!" echo "" fi # ============================================================================= # CONFIGURATION HELPERS: Configure settings when needed by phases # ============================================================================= configure_basic_settings() { if [ ! -f "${WC_HOME}/config.yaml" ] || [ -z "$(get_current_config "operator.email")" ]; then print_header "Basic Configuration" # Detect current network for suggestions CURRENT_IP=$(ip route get 8.8.8.8 | awk '{print $7; exit}' 2>/dev/null || echo "192.168.1.100") GATEWAY_IP=$(ip route | grep default | awk '{print $3; exit}' 2>/dev/null || echo "192.168.1.1") SUBNET_PREFIX=$(echo "${CURRENT_IP}" | cut -d. -f1-3) print_info "Detected network: ${SUBNET_PREFIX}.x (gateway: ${GATEWAY_IP})" echo "This will configure basic settings for your wild-cloud deployment." echo "" # Basic Information current_email=$(get_current_config "operator.email") email=$(prompt_with_default "Your email address (for Let's Encrypt certificates)" "" "${current_email}") wild-config-set "operator.email" "${email}" # Domain Configuration current_base_domain=$(get_current_config "cloud.baseDomain") base_domain=$(prompt_with_default "Your base domain name (e.g., example.com)" "" "${current_base_domain}") wild-config-set "cloud.baseDomain" "${base_domain}" current_domain=$(get_current_config "cloud.domain") domain=$(prompt_with_default "Your public cloud domain" "cloud.${base_domain}" "${current_domain}") wild-config-set "cloud.domain" "${domain}" current_internal_domain=$(get_current_config "cloud.internalDomain") internal_domain=$(prompt_with_default "Your internal cloud domain" "internal.${domain}" "${current_internal_domain}") wild-config-set "cloud.internalDomain" "${internal_domain}" # Derive cluster name from domain cluster_name=$(echo "${domain}" | tr '.' '-' | tr '[:upper:]' '[:lower:]') wild-config-set "cluster.name" "${cluster_name}" print_info "Set cluster name to: ${cluster_name}" print_success "Basic configuration completed" echo "" fi } configure_dns_and_certificates() { if [ -z "$(get_current_config "cluster.certManager.cloudflare.domain")" ]; then print_header "DNS and Certificate Configuration" echo "For automatic SSL certificates and DNS management, we use Cloudflare." echo "" base_domain=$(get_current_config "cloud.baseDomain") domain=$(get_current_config "cloud.domain") echo "Is your domain '${base_domain}' registered and managed through Cloudflare? (y/n)" read -r use_cloudflare if [[ "${use_cloudflare}" =~ ^[Yy]$ ]]; then wild-config-set "cluster.certManager.cloudflare.domain" "${domain}" current_cf_token=$(get_current_secret "cloudflare.token") if [ -z "${current_cf_token}" ]; then echo "" print_info "You'll need a Cloudflare API token with the following permissions:" echo " - Zone:Zone:Read" echo " - Zone:DNS:Edit" echo " - Include:All zones" echo "" echo "Create one at: https://dash.cloudflare.com/profile/api-tokens" echo "" fi cf_token=$(prompt_with_default "Cloudflare API token" "" "${current_cf_token}") wild-secret-set "cloudflare.token" "${cf_token}" else print_warning "You'll need to configure DNS and SSL certificates manually." print_info "Consider transferring your domain to Cloudflare for easier management." fi print_success "DNS and certificate configuration completed" echo "" fi } configure_network_settings() { if [ -z "$(get_current_config "cloud.router.ip")" ]; then print_header "Network Configuration" CURRENT_IP=$(ip route get 8.8.8.8 | awk '{print $7; exit}' 2>/dev/null || echo "192.168.1.100") GATEWAY_IP=$(ip route | grep default | awk '{print $3; exit}' 2>/dev/null || echo "192.168.1.1") SUBNET_PREFIX=$(echo "${CURRENT_IP}" | cut -d. -f1-3) current_router_ip=$(get_current_config "cloud.router.ip") router_ip=$(prompt_with_default "Router/Gateway IP" "${GATEWAY_IP}" "${current_router_ip}") wild-config-set "cloud.router.ip" "${router_ip}" current_dns_ip=$(get_current_config "cloud.dns.ip") dns_ip=$(prompt_with_default "DNS server IP (dnsmasq machine)" "${SUBNET_PREFIX}.50" "${current_dns_ip}") wild-config-set "cloud.dns.ip" "${dns_ip}" current_dhcp_range=$(get_current_config "cloud.dhcpRange") dhcp_range=$(prompt_with_default "DHCP range for dnsmasq" "${SUBNET_PREFIX}.100,${SUBNET_PREFIX}.200" "${current_dhcp_range}") wild-config-set "cloud.dhcpRange" "${dhcp_range}" current_interface=$(get_current_config "cloud.dnsmasq.interface") interface=$(prompt_with_default "Network interface for dnsmasq" "eth0" "${current_interface}") wild-config-set "cloud.dnsmasq.interface" "${interface}" current_external_resolver=$(get_current_config "cloud.dns.externalResolver") external_resolver=$(prompt_with_default "External DNS resolver" "1.1.1.1" "${current_external_resolver}") wild-config-set "cloud.dns.externalResolver" "${external_resolver}" print_success "Network configuration completed" echo "" fi } configure_cluster_settings() { if [ -z "$(get_current_config "cluster.nodes.talos.version")" ]; then print_header "Kubernetes Cluster Configuration" CURRENT_IP=$(ip route get 8.8.8.8 | awk '{print $7; exit}' 2>/dev/null || echo "192.168.1.100") SUBNET_PREFIX=$(echo "${CURRENT_IP}" | cut -d. -f1-3) current_talos_version=$(get_current_config "cluster.nodes.talos.version") talos_version=$(prompt_with_default "Talos version" "v1.6.1" "${current_talos_version}") wild-config-set "cluster.nodes.talos.version" "${talos_version}" current_ip_pool=$(get_current_config "cluster.ipAddressPool") ip_pool=$(prompt_with_default "MetalLB IP address pool" "${SUBNET_PREFIX}.80-${SUBNET_PREFIX}.89" "${current_ip_pool}") wild-config-set "cluster.ipAddressPool" "${ip_pool}" # Automatically set load balancer IP to first address in the pool lb_ip=$(echo "${ip_pool}" | cut -d'-' -f1) wild-config-set "cluster.loadBalancerIp" "${lb_ip}" print_info "Set load balancer IP to: ${lb_ip} (first IP in MetalLB pool)" # Control plane nodes echo "" print_info "Configure control plane nodes (you need at least 3 for HA):" current_vip=$(get_current_config "cluster.nodes.control.vip") vip=$(prompt_with_default "Control plane virtual IP" "${SUBNET_PREFIX}.90" "${current_vip}") wild-config-set "cluster.nodes.control.vip" "${vip}" for i in 1 2 3; do current_node_ip=$(get_current_config "cluster.nodes.control.node${i}.ip") node_ip=$(prompt_with_default "Control plane node ${i} IP address" "${SUBNET_PREFIX}.$(( 90 + i ))" "${current_node_ip}") wild-config-set "cluster.nodes.control.node${i}.ip" "${node_ip}" done # Talos schematic ID current_schematic_id=$(get_current_config "cluster.nodes.talos.schematicId") echo "" print_info "Get your Talos schematic ID from: https://factory.talos.dev/" print_info "This customizes Talos with the drivers needed for your hardware." schematic_id=$(prompt_with_default "Talos schematic ID" "" "${current_schematic_id}") wild-config-set "cluster.nodes.talos.schematicId" "${schematic_id}" # External DNS cluster_name=$(get_current_config "cluster.name") current_owner_id=$(get_current_config "cluster.externalDns.ownerId") owner_id=$(prompt_with_default "External DNS owner ID" "external-dns-${cluster_name}" "${current_owner_id}") wild-config-set "cluster.externalDns.ownerId" "${owner_id}" print_success "Cluster configuration completed" echo "" fi } configure_storage_settings() { if [ -z "$(get_current_config "cloud.nfs.host")" ]; then print_header "Storage Configuration" dns_ip=$(get_current_config "cloud.dns.ip") internal_domain=$(get_current_config "cloud.internalDomain") current_nfs_host=$(get_current_config "cloud.nfs.host") nfs_host=$(prompt_with_default "NFS server host" "${dns_ip}" "${current_nfs_host}") wild-config-set "cloud.nfs.host" "${nfs_host}" current_media_path=$(get_current_config "cloud.nfs.mediaPath") media_path=$(prompt_with_default "NFS media path" "/mnt/storage/media" "${current_media_path}") wild-config-set "cloud.nfs.mediaPath" "${media_path}" current_storage_capacity=$(get_current_config "cloud.nfs.storageCapacity") storage_capacity=$(prompt_with_default "Storage capacity for NFS PV" "1Ti" "${current_storage_capacity}") wild-config-set "cloud.nfs.storageCapacity" "${storage_capacity}" # Docker Registry current_registry_host=$(get_current_config "cloud.dockerRegistryHost") registry_host=$(prompt_with_default "Docker registry hostname" "registry.${internal_domain}" "${current_registry_host}") wild-config-set "cloud.dockerRegistryHost" "${registry_host}" print_success "Storage configuration completed" echo "" fi } # ============================================================================= # HELPER FUNCTION: Copy setup files on demand # ============================================================================= copy_setup_files_if_needed() { local setup_type="$1" # "cluster-nodes" or "cluster" SOURCE_DIR="${WC_ROOT}/setup" DEST_DIR="${WC_HOME}/setup" if [ ! -d "${DEST_DIR}/${setup_type}" ]; then print_info "Copying ${setup_type} setup files..." mkdir -p "${DEST_DIR}" cp -r "${SOURCE_DIR}/${setup_type}" "${DEST_DIR}/${setup_type}" # Copy README if it doesn't exist if [ ! -f "${DEST_DIR}/README.md" ]; then cp "${SOURCE_DIR}/README.md" "${DEST_DIR}/README.md" fi print_success "${setup_type} setup files copied" fi } # ============================================================================= # PHASE 1: Installer Image Generation # ============================================================================= if [ "${SKIP_INSTALLER}" = false ]; then print_header "Phase 1: Installer Image Generation" # Configure basic settings and cluster settings if needed configure_basic_settings configure_cluster_settings # Get Talos version and schematic ID from config TALOS_VERSION=$(wild-config cluster.nodes.talos.version) SCHEMATIC_ID=$(wild-config cluster.nodes.talos.schematicId) print_info "Creating custom Talos installer image..." print_info "Talos version: $TALOS_VERSION" # Check if schematic ID exists if [ -z "$SCHEMATIC_ID" ] || [ "$SCHEMATIC_ID" = "null" ]; then print_error "No schematic ID found in config.yaml" print_info "You can get a schematic ID from: https://factory.talos.dev/" read -p "Enter schematic ID: " -r SCHEMATIC_ID if [ -n "$SCHEMATIC_ID" ]; then wild-config-set "cluster.nodes.talos.schematicId" "$SCHEMATIC_ID" else print_error "Schematic ID required for installer image generation" exit 1 fi fi print_info "Schematic ID: $SCHEMATIC_ID" if [ -f "${WC_HOME}/config.yaml" ] && yq eval '.cluster.nodes.talos.schematic.customization.systemExtensions.officialExtensions' "${WC_HOME}/config.yaml" >/dev/null 2>&1; then echo "" print_info "Schematic includes:" yq eval '.cluster.nodes.talos.schematic.customization.systemExtensions.officialExtensions[]' "${WC_HOME}/config.yaml" | sed 's/^/ - /' || true echo "" fi # Generate installer image URL INSTALLER_URL="factory.talos.dev/metal-installer/$SCHEMATIC_ID:$TALOS_VERSION" print_success "Custom installer image URL generated!" print_info "Installer URL: $INSTALLER_URL" print_success "Phase 1 completed: Installer image generated" echo "" else print_info "Skipping Phase 1: Installer Image Generation" fi # ============================================================================= # PHASE 2: Node Hardware Detection (from detect-node-hardware.sh) # ============================================================================= if [ "${SKIP_HARDWARE}" = false ]; then print_header "Phase 2: Node Hardware Detection" # Configure basic settings and cluster settings if needed configure_basic_settings configure_cluster_settings print_info "This phase will help you register Talos nodes by discovering their hardware." print_info "You'll need nodes booted in maintenance mode and accessible via IP." echo "" # Register up to 3 control plane nodes for i in 1 2 3; do echo "" print_info "Configure control plane node $i:" read -p "Do you want to register control plane node $i now? (y/N): " -r register_node if [[ $register_node =~ ^[Yy]$ ]]; then read -p "Enter maintenance IP for node $i: " -r NODE_IP if [ -z "$NODE_IP" ]; then print_warning "Skipping node $i registration" continue fi print_info "Registering Talos control plane node $i at $NODE_IP..." # Test connectivity print_info "Testing connectivity to node..." if ! talosctl -n "$NODE_IP" get links --insecure >/dev/null 2>&1; then print_error "Cannot connect to node at $NODE_IP" print_info "Make sure the node is booted in maintenance mode and accessible." continue fi print_success "Node is accessible" # Discover network interfaces print_info "Discovering network interfaces..." # Find the interface with default route CONNECTED_INTERFACE=$(talosctl -n "$NODE_IP" get routes --insecure -o json 2>/dev/null | \ jq -s -r '.[] | select(.spec.destination == "0.0.0.0/0" and .spec.gateway != null) | .spec.outLinkName' | \ head -1) if [ -n "$CONNECTED_INTERFACE" ]; then ACTIVE_INTERFACE="$CONNECTED_INTERFACE" print_success "Discovered connected interface (with default route): $ACTIVE_INTERFACE" else # Fallback: find any active ethernet interface print_info "No default route found, checking for active ethernet interfaces..." ACTIVE_INTERFACE=$(talosctl -n "$NODE_IP" get links --insecure -o json 2>/dev/null | \ jq -s -r '.[] | select(.spec.operationalState == "up" and .spec.type == "ether" and .metadata.id != "lo") | .metadata.id' | \ head -1) if [ -z "$ACTIVE_INTERFACE" ]; then print_error "No active ethernet interface found" print_info "Available interfaces:" talosctl -n "$NODE_IP" get links --insecure continue fi print_success "Discovered active interface: $ACTIVE_INTERFACE" fi # Discover available disks print_info "Discovering available disks..." AVAILABLE_DISKS=$(talosctl -n "$NODE_IP" get disks --insecure -o json 2>/dev/null | \ jq -s -r '.[] | select(.spec.size > 10000000000) | .metadata.id' | \ head -5) if [ -z "$AVAILABLE_DISKS" ]; then print_error "No suitable disks found (must be >10GB)" print_info "Available disks:" talosctl -n "$NODE_IP" get disks --insecure continue fi print_info "Available disks (>10GB):" echo "$AVAILABLE_DISKS" echo "" # Let user choose disk print_info "Select installation disk for node $i:" select INSTALL_DISK in $AVAILABLE_DISKS; do if [ -n "${INSTALL_DISK:-}" ]; then break fi echo "Invalid selection. Please try again." done # Add /dev/ prefix if not present if [[ "$INSTALL_DISK" != /dev/* ]]; then INSTALL_DISK="/dev/$INSTALL_DISK" fi print_success "Selected disk: $INSTALL_DISK" # Update config.yaml with per-node configuration print_info "Updating config.yaml with node $i configuration..." CONFIG_FILE="${WC_HOME}/config.yaml" # Get the target IP for this node from the existing config TARGET_IP=$(yq eval ".cluster.nodes.control.node${i}.ip" "$CONFIG_FILE") # Use yq to update the per-node configuration yq eval ".cluster.nodes.control.node${i}.ip = \"$TARGET_IP\"" -i "$CONFIG_FILE" yq eval ".cluster.nodes.control.node${i}.interface = \"$ACTIVE_INTERFACE\"" -i "$CONFIG_FILE" yq eval ".cluster.nodes.control.node${i}.disk = \"$INSTALL_DISK\"" -i "$CONFIG_FILE" print_success "Updated config.yaml for node $i:" print_info " - Target IP: $TARGET_IP" print_info " - Network interface: $ACTIVE_INTERFACE" print_info " - Installation disk: $INSTALL_DISK" else print_info "Skipping node $i registration" fi done print_success "Phase 2 completed: Node hardware detection" echo "" else print_info "Skipping Phase 2: Node Hardware Detection" fi # ============================================================================= # PHASE 3: Machine Config Generation (from generate-machine-configs.sh) # ============================================================================= if [ "${SKIP_CONFIGS}" = false ]; then print_header "Phase 3: Machine Config Generation" # Configure basic settings and cluster settings if needed configure_basic_settings configure_cluster_settings # Copy cluster-nodes setup files if needed for this phase copy_setup_files_if_needed "cluster-nodes" NODE_SETUP_DIR="${WC_HOME}/setup/cluster-nodes" # Check if cluster has been initialized if [ ! -f "${NODE_SETUP_DIR}/generated/secrets.yaml" ]; then print_error "Cluster not initialized. You need to run cluster initialization first." print_info "This typically involves running talosctl gen config to generate initial secrets." read -p "Do you want to generate initial cluster secrets now? (y/N): " -r generate_secrets if [[ $generate_secrets =~ ^[Yy]$ ]]; then # Generate cluster secrets CLUSTER_NAME=$(wild-config cluster.name) VIP=$(wild-config cluster.nodes.control.vip) print_info "Generating initial cluster configuration..." mkdir -p "${NODE_SETUP_DIR}/generated" cd "${NODE_SETUP_DIR}/generated" talosctl gen config "$CLUSTER_NAME" "https://$VIP:6443" cd - >/dev/null print_success "Initial cluster configuration generated" else print_warning "Skipping machine config generation - cluster secrets required" SKIP_CONFIGS=true fi fi if [ "${SKIP_CONFIGS}" = false ]; then # Get cluster configuration from config.yaml CLUSTER_NAME=$(wild-config cluster.name) VIP=$(wild-config cluster.nodes.control.vip) print_info "Generating machine configurations for cluster: $CLUSTER_NAME" # Check which nodes have been registered (have hardware config) REGISTERED_NODES=() for i in 1 2 3; do if yq eval ".cluster.nodes.control.node${i}.interface" "${WC_HOME}/config.yaml" | grep -v "null" >/dev/null 2>&1; then NODE_IP=$(wild-config cluster.nodes.control.node${i}.ip) REGISTERED_NODES+=("$NODE_IP") print_success "Node $i registered: $NODE_IP" else print_info "Node $i not registered yet" fi done if [ ${#REGISTERED_NODES[@]} -eq 0 ]; then print_warning "No nodes have been registered yet." print_info "You can register nodes in Phase 4 or run detect-node-hardware.sh separately." else # Create directories mkdir -p "${NODE_SETUP_DIR}/final" "${NODE_SETUP_DIR}/patch" # Compile patch templates for registered nodes only print_info "Compiling patch templates..." for i in 1 2 3; do if yq eval ".cluster.nodes.control.node${i}.interface" "${WC_HOME}/config.yaml" | grep -v "null" >/dev/null 2>&1; then print_info "Compiling template for control plane node $i..." cat "${NODE_SETUP_DIR}/patch.templates/controlplane-node-${i}.yaml" | wild-compile-template > "${NODE_SETUP_DIR}/patch/controlplane-node-${i}.yaml" fi done # Always compile worker template (doesn't require hardware detection) if [ -f "${NODE_SETUP_DIR}/patch.templates/worker.yaml" ]; then cat "${NODE_SETUP_DIR}/patch.templates/worker.yaml" | wild-compile-template > "${NODE_SETUP_DIR}/patch/worker.yaml" fi # Generate final machine configs for registered nodes only print_info "Generating final machine configurations..." for i in 1 2 3; do if yq eval ".cluster.nodes.control.node${i}.interface" "${WC_HOME}/config.yaml" | grep -v "null" >/dev/null 2>&1; then print_info "Generating config for control plane node $i..." talosctl machineconfig patch "${NODE_SETUP_DIR}/generated/controlplane.yaml" --patch @"${NODE_SETUP_DIR}/patch/controlplane-node-${i}.yaml" -o "${NODE_SETUP_DIR}/final/controlplane-node-${i}.yaml" fi done # Always generate worker config (doesn't require hardware detection) if [ -f "${NODE_SETUP_DIR}/patch/worker.yaml" ]; then print_info "Generating worker config..." talosctl machineconfig patch "${NODE_SETUP_DIR}/generated/worker.yaml" --patch @"${NODE_SETUP_DIR}/patch/worker.yaml" -o "${NODE_SETUP_DIR}/final/worker.yaml" fi # Update talosctl context with registered nodes print_info "Updating talosctl context..." if [ ${#REGISTERED_NODES[@]} -gt 0 ]; then talosctl config node "${REGISTERED_NODES[@]}" fi print_success "Machine configurations generated successfully!" echo "" print_info "Generated configs:" for i in 1 2 3; do if [ -f "${NODE_SETUP_DIR}/final/controlplane-node-${i}.yaml" ]; then NODE_IP=$(wild-config cluster.nodes.control.node${i}.ip) print_info " - ${NODE_SETUP_DIR}/final/controlplane-node-${i}.yaml (target IP: $NODE_IP)" fi done if [ -f "${NODE_SETUP_DIR}/final/worker.yaml" ]; then print_info " - ${NODE_SETUP_DIR}/final/worker.yaml" fi fi fi print_success "Phase 3 completed: Machine config generation" echo "" else print_info "Skipping Phase 3: Machine Config Generation" fi # ============================================================================= # PHASE 4: Cluster Services Installation (from install-all.sh) # ============================================================================= if [ "${SKIP_INSTALL}" = false ]; then print_header "Phase 4: Cluster Services Installation" # Configure settings needed for cluster services configure_basic_settings configure_dns_and_certificates configure_network_settings configure_storage_settings # Copy cluster services setup files if needed for this phase copy_setup_files_if_needed "cluster" print_info "This phase installs core cluster services (MetalLB, Traefik, cert-manager, etc.)" print_warning "Make sure your cluster is running and kubectl is configured!" read -p "Do you want to install cluster services now? (y/N): " -r install_services if [[ $install_services =~ ^[Yy]$ ]]; then # Check if kubectl works if ! kubectl cluster-info >/dev/null 2>&1; then print_error "kubectl is not configured or cluster is not accessible" print_info "Make sure your cluster is running and kubeconfig is set up" print_info "You can get kubeconfig with: talosctl kubeconfig" SKIP_INSTALL=true else print_info "Installing cluster services..." CLUSTER_SETUP_DIR="${WC_HOME}/setup/cluster" if [ ! -d "$CLUSTER_SETUP_DIR" ]; then print_error "Cluster setup directory not found: $CLUSTER_SETUP_DIR" print_info "Make sure Phase 2 (Setup Template Copying) has been completed" SKIP_INSTALL=true else cd "$CLUSTER_SETUP_DIR" print_info "Installing MetalLB..." if [ -f "./metallb/install.sh" ]; then ./metallb/install.sh else print_warning "MetalLB install script not found" fi print_info "Installing Longhorn..." if [ -f "./longhorn/install.sh" ]; then ./longhorn/install.sh else print_warning "Longhorn install script not found" fi print_info "Installing Traefik..." if [ -f "./traefik/install.sh" ]; then ./traefik/install.sh else print_warning "Traefik install script not found" fi print_info "Installing CoreDNS..." if [ -f "./coredns/install.sh" ]; then ./coredns/install.sh else print_warning "CoreDNS install script not found" fi print_info "Installing cert-manager..." if [ -f "./cert-manager/install.sh" ]; then ./cert-manager/install.sh else print_warning "cert-manager install script not found" fi print_info "Installing ExternalDNS..." if [ -f "./externaldns/install.sh" ]; then ./externaldns/install.sh else print_warning "ExternalDNS install script not found" fi print_info "Installing Kubernetes Dashboard..." if [ -f "./kubernetes-dashboard/install.sh" ]; then ./kubernetes-dashboard/install.sh else print_warning "Kubernetes Dashboard install script not found" fi print_info "Installing NFS..." if [ -f "./nfs/install.sh" ]; then ./nfs/install.sh else print_warning "NFS install script not found" fi print_info "Installing Docker Registry..." if [ -f "./docker-registry/install.sh" ]; then ./docker-registry/install.sh else print_warning "Docker Registry install script not found" fi cd - >/dev/null print_success "Infrastructure setup complete!" fi fi else print_info "Skipping cluster services installation" SKIP_INSTALL=true fi if [ "${SKIP_INSTALL}" = false ]; then print_success "Phase 4 completed: Cluster services installation" fi echo "" else print_info "Skipping Phase 4: Cluster Services Installation" fi # ============================================================================= # FINAL SUMMARY # ============================================================================= print_header "Wild-Cloud Setup Complete!" print_success "All phases completed successfully!" echo "" print_info "What was accomplished:" print_info "✅ Cloud setup completed" if [ "${SKIP_INSTALLER}" = false ]; then print_info "✅ Phase 1: Installer image generated" else print_info "⏸️ Phase 1: Installer image generation (skipped)" fi if [ "${SKIP_HARDWARE}" = false ]; then print_info "✅ Phase 2: Node hardware detection completed" else print_info "⏸️ Phase 2: Node hardware detection (skipped)" fi if [ "${SKIP_CONFIGS}" = false ]; then print_info "✅ Phase 3: Machine configs generated" else print_info "⏸️ Phase 3: Machine config generation (skipped)" fi if [ "${SKIP_INSTALL}" = false ]; then print_info "✅ Phase 4: Cluster services installed" else print_info "⏸️ Phase 4: Cluster services installation (skipped)" fi print_info "✅ Configuration completed as needed by phases" echo "" print_info "Configuration files:" echo " - ${WC_HOME}/config.yaml" echo " - ${WC_HOME}/secrets.yaml" if [ -d "${WC_HOME}/setup/cluster-nodes/final" ] && [ "$(ls -A ${WC_HOME}/setup/cluster-nodes/final 2>/dev/null)" ]; then echo "" print_info "Machine configurations:" for config_file in "${WC_HOME}/setup/cluster-nodes/final"/*.yaml; do if [ -f "$config_file" ]; then echo " - $config_file" fi done fi echo "" print_info "Next steps:" echo " 1. Review your configuration and generated files" if [ "${SKIP_HARDWARE}" = true ] || [ "${SKIP_CONFIGS}" = true ]; then echo " 2. Complete any skipped phases as needed:" if [ "${SKIP_HARDWARE}" = true ]; then echo " - Re-run wild-init to continue with hardware detection" fi if [ "${SKIP_CONFIGS}" = true ]; then echo " - Generate machine configs after hardware detection" fi fi if [ "${SKIP_INSTALL}" = false ] && command -v kubectl >/dev/null 2>&1; then INTERNAL_DOMAIN=$(wild-config cloud.internalDomain 2>/dev/null || echo "your-internal-domain") echo " 2. Access the dashboard at: https://dashboard.${INTERNAL_DOMAIN}" echo " 3. Get the dashboard token with: ./bin/dashboard-token" echo "" echo "To verify components, run:" echo " - kubectl get pods -n cert-manager" echo " - kubectl get pods -n externaldns" echo " - kubectl get pods -n kubernetes-dashboard" echo " - kubectl get clusterissuers" else echo " 2. Set up your cluster and install services" echo " 3. Apply machine configurations to your nodes" fi echo ""