fix: use navidrome-v8 and pin to opti1 for migration

.github/workflows/deploy.yml

@@ -23,9 +23,13 @@ jobs:
         uses: actions/checkout@v4
         
       - name: Setup Nomad CLI
-        uses: hashicorp/setup-nomad@v2
-        with:
-          version: '1.10.5'
+        run: |
+          NOMAD_VERSION="1.10.5"
+          curl -sSL https://releases.hashicorp.com/nomad/${NOMAD_VERSION}/nomad_${NOMAD_VERSION}_linux_amd64.zip -o nomad.zip
+          unzip nomad.zip
+          sudo mv nomad /usr/local/bin/
+          rm nomad.zip
+          nomad version
 
       - name: Set Container Version
         id: container_version

Dockerfile

@@ -18,6 +18,9 @@ RUN chmod +x /usr/local/bin/entrypoint.sh
 # Copy LiteFS configuration
 COPY litefs.yml /etc/litefs.yml
 
+# Create mount points and data directories
+RUN mkdir -p /litefs /data
+
 # LiteFS becomes the supervisor.
 
 # It will mount the FUSE fs and then execute the command defined in litefs.yml's exec section.

entrypoint.sh

@@ -1,14 +1,10 @@
 #!/bin/bash
-set -e
-
 # Configuration from environment
 SERVICE_NAME="navidrome"
-# Use Nomad allocation ID for a unique service ID
 SERVICE_ID="${SERVICE_NAME}-${NOMAD_ALLOC_ID:-$(hostname)}"
 PORT=4533
 CONSUL_HTTP_ADDR="${CONSUL_URL:-http://localhost:8500}"
 NODE_IP="${ADVERTISE_IP}"
-DB_LOCK_FILE="/data/.primary"
 NAVIDROME_PID=0
 
 # Tags for the Primary service (Traefik enabled)
@@ -16,29 +12,43 @@ PRIMARY_TAGS='["navidrome","web","traefik.enable=true","urlprefix-/navidrome","t
 
 # --- Helper Functions ---
 
-# Backup Database (Only on Primary)
-run_backup() {
-  local backup_dir="/shared_data/backup"
-  local timestamp=$(date +%Y%m%d_%H%M%S)
-  local backup_file="${backup_dir}/navidrome.db_${timestamp}.bak"
+# Check if this node is the LiteFS Primary
+check_primary() {
+  local status=$(curl -s http://localhost:20202/info || echo "{}")
+  local is_primary=$(echo "$status" | jq -r 'if type == "object" then (.isPrimary // false) else false end' 2>/dev/null || echo "false")
   
-  echo "Backing up database to ${backup_file}..."
-  mkdir -p "$backup_dir"
-  
-  if litefs export -name navidrome.db "$backup_file"; then
-    echo "Backup successful."
-    # Keep only last 7 days
-    find "$backup_dir" -name "navidrome.db_*.bak" -mtime +7 -delete
-    echo "Old backups cleaned."
-  else
-    echo "ERROR: Backup failed!"
+  if [ "$is_primary" = "true" ]; then
+    return 0 # We are the primary
   fi
+  return 1 # We are a replica
+}
+
+# Wait for LiteFS to settle and determine its role
+wait_for_litefs() {
+  echo "Waiting for LiteFS to settle..."
+  local timeout=60
+  local count=0
+  while [ $count -lt $timeout ]; do
+    local status=$(curl -s http://localhost:20202/info || echo "null")
+    local is_primary_val=$(echo "$status" | jq -r 'if type == "object" then (.isPrimary // "null") else "null" end' 2>/dev/null || echo "null")
+    
+    if [ "$is_primary_val" != "null" ]; then
+      local role="replica"
+      if [ "$is_primary_val" = "true" ]; then role="primary"; fi
+      echo "LiteFS initialized. Role: $role"
+      return 0
+    fi
+    sleep 2
+    count=$((count + 2))
+    echo -n "."
+  done
+  echo "ERROR: LiteFS failed to settle after ${timeout}s"
+  return 1
 }
 
 # Register Service with TTL Check
 register_service() {
-  echo "Promoted! Registering service ${SERVICE_ID}..."
-  # Convert bash list string to JSON array if needed, but PRIMARY_TAGS is already JSON-like
+  echo "Registering service ${SERVICE_ID} with Consul..."
   curl -s -X PUT "${CONSUL_HTTP_ADDR}/v1/agent/service/register" -d "{
     \"ID\": \"${SERVICE_ID}\",
     \"Name\": \"${SERVICE_NAME}\",
@@ -59,7 +69,7 @@ pass_ttl() {
 
 # Deregister Service
 deregister_service() {
-  echo "Demoted/Stopping. Deregistering service ${SERVICE_ID}..."
+  echo "Deregistering service ${SERVICE_ID} from Consul..."
   curl -s -X PUT "${CONSUL_HTTP_ADDR}/v1/agent/service/deregister/${SERVICE_ID}"
 }
 
@@ -68,11 +78,48 @@ start_app() {
   echo "Node is Primary. Starting Navidrome..."
   
   # Ensure shared directories exist
-  mkdir -p /shared_data/plugins /shared_data/cache /shared_data/backup
+  mkdir -p /shared_data/plugins /shared_data/cache /shared_data/backup /shared_data/artist_images /shared_data/artwork
+
+  # SEEDING LOGIC: If DB doesn't exist in cluster, restore from backup
+  if [ ! -f /data/navidrome.db ]; then
+    echo "Database /data/navidrome.db not found. Looking for backups to seed..."
+    local latest_backup=$(ls -t /shared_data/backup/navidrome.db_*.bak 2>/dev/null | head -n 1)
+    if [ -n "$latest_backup" ]; then
+       echo "Seeding from $latest_backup..."
+       litefs import -name navidrome.db "$latest_backup"
+    else
+       echo "No backups found. Navidrome will start with a fresh database."
+    fi
+  fi
+
+  # Wait for LiteFS to expose the DB file
+  echo "Waiting for /data/navidrome.db to be exposed by LiteFS..."
+  local db_timeout=30
+  local db_count=0
+  while [ ! -f /data/navidrome.db ] && [ $db_count -lt $db_timeout ]; do
+    sleep 1
+    db_count=$((db_count + 1))
+  done
+
+  # Setup local data folder with BIND MOUNT for the DB
+  # This allows SQLite to create -wal/-shm files in the local writable directory
+  # while the main DB file is managed by LiteFS.
+  rm -rf /local/navidrome_data
+  mkdir -p /local/navidrome_data
+  
+  touch /local/navidrome_data/navidrome.db
+  mount --bind /data/navidrome.db /local/navidrome_data/navidrome.db
+
+  # Configuration
+  export ND_DATAFOLDER="/local/navidrome_data"
+  export ND_CACHEFOLDER="/shared_data/cache"
+  export ND_BACKUP_PATH="/shared_data/backup"
+  export ND_PLUGINS_FOLDER="/shared_data/plugins"
+  export ND_ARTISTIMAGEFOLDER="artist_images"
 
   /app/navidrome &
   NAVIDROME_PID=$!
-  echo "Navidrome started with PID ${NAVIDROME_PID}"
+  echo "Navidrome running (PID: $NAVIDROME_PID) with DataFolder at /local/navidrome_data (DB bind-mounted)"
 }
 
 # Stop Navidrome
@@ -82,13 +129,13 @@ stop_app() {
     kill -SIGTERM "${NAVIDROME_PID}"
     wait "${NAVIDROME_PID}" 2>/dev/null || true
     NAVIDROME_PID=0
+    umount /local/navidrome_data/navidrome.db 2>/dev/null || true
   fi
 }
 
-# --- Signal Handling (The Safety Net) ---
-# If Nomad stops the container, we stop the app and deregister.
+# --- Cleanup ---
 cleanup() {
-  echo "Caught signal, shutting down..."
+  echo "Shutting down..."
   stop_app
   deregister_service
   exit 0
@@ -99,55 +146,23 @@ trap cleanup TERM INT
 # --- Main Loop ---
 
 echo "Starting Supervisor. Waiting for leadership settle..."
-echo "Node IP: $NODE_IP"
-echo "Consul: $CONSUL_HTTP_ADDR"
-
-# Small sleep to let LiteFS settle and leadership election complete
-sleep 5
-
-LAST_BACKUP_TIME=0
-BACKUP_INTERVAL=86400 # 24 hours
+wait_for_litefs || exit 1
 
 while true; do
-  # In LiteFS 0.5, .primary file exists ONLY on replicas.
-  if [ ! -f "$DB_LOCK_FILE" ]; then
+  if check_primary; then
     # === WE ARE PRIMARY ===
-    
-    # 1. If App is not running, start it and register
     if [ "${NAVIDROME_PID}" -eq 0 ] || ! kill -0 "${NAVIDROME_PID}" 2>/dev/null; then
-      if [ "${NAVIDROME_PID}" -gt 0 ]; then
-        echo "CRITICAL: Navidrome crashed! Restarting..."
-      fi
       start_app
       register_service
     fi
-
-    # 2. Maintain the heartbeat (TTL)
     pass_ttl
-    
-    # 3. Handle periodic backup
-    CURRENT_TIME=$(date +%s)
-    if [ $((CURRENT_TIME - LAST_BACKUP_TIME)) -ge $BACKUP_INTERVAL ]; then
-      run_backup
-      LAST_BACKUP_TIME=$CURRENT_TIME
-    fi
-    
   else
     # === WE ARE REPLICA ===
-    
-    # If App is running (we were just demoted), stop it
     if [ "${NAVIDROME_PID}" -gt 0 ]; then
       echo "Lost leadership. Demoting..."
       stop_app
       deregister_service
-      # Reset backup timer so the next primary can start fresh or we start fresh if promoted again
-      LAST_BACKUP_TIME=0
     fi
-    
-    # No service registration exists for replicas to keep Consul clean.
   fi
-
-  # Sleep short enough to update TTL (every 5s is safe for 15s TTL)
-  sleep 5 &
-  wait $! # Wait allows the 'trap' to interrupt the sleep instantly
+  sleep 10
 done

litefs.yml

@@ -8,29 +8,19 @@ data:
 # Use Consul for leader election
 lease:
   type: "consul"
+  candidate: true
+  promote: true
   advertise-url: "http://${ADVERTISE_IP}:20202"
   consul:
     url: "${CONSUL_URL}"
-    key: "litefs/navidrome"
+    key: "litefs/navidrome-v8"
+    ttl: "30s"
+    lock-delay: "5s"
 
 # Internal HTTP API for replication
 http:
   addr: "0.0.0.0:20202"
 
-# The HTTP Proxy routes traffic to handle write-forwarding
-proxy:
-  addr: ":8080"
-  target: "localhost:4533"
-  db: "navidrome.db"
-  passthrough:
-    - "*.js"
-    - "*.css"
-    - "*.png"
-    - "*.jpg"
-    - "*.jpeg"
-    - "*.gif"
-    - "*.svg"
-
 # Commands to run only on the primary node. 
 exec:
   - cmd: "/usr/local/bin/entrypoint.sh"

navidrome-litefs.nomad

@@ -1,12 +1,12 @@
+variable "container_sha" {
+  type    = string
+  default = "045fc6e82b9ecb6bebc1f095f62498935df70bbf"
+}
+
 job "navidrome-litefs" {
   datacenters = ["dc1"]
   type        = "service"
 
-  variable "container_sha" {
-    type    = string
-    default = "045fc6e82b9ecb6bebc1f095f62498935df70bbf"
-  }
-
   constraint {
     attribute = "${attr.kernel.name}"
     value     = "linux"
@@ -63,10 +63,11 @@ job "navidrome-litefs" {
         PORT         = "8080" # Internal proxy port (unused but kept)
 
         # Navidrome Config
-        ND_DATAFOLDER    = "/data"
+        ND_DATAFOLDER     = "/shared_data"
         ND_PLUGINS_FOLDER = "/shared_data/plugins"
-        ND_CACHEFOLDER   = "/shared_data/cache"
-        ND_BACKUP_PATH   = "/shared_data/backup"
+        ND_CACHEFOLDER    = "/shared_data/cache"
+        ND_BACKUP_PATH    = "/shared_data/backup"
+        ND_ARTISTIMAGEFOLDER = "artist_images"
         ND_BACKUPSCHEDULE = ""
         
         ND_SCANSCHEDULE              = "0"