feat(entrypoint): Refactor entrypoint for leadership-aware process management and clean up register.sh

Dockerfile

@@ -12,9 +12,8 @@ RUN apk add --no-cache fuse3 ca-certificates bash curl
 COPY --from=litefs /usr/local/bin/litefs /usr/local/bin/litefs
 
 # Copy scripts
-COPY register.sh /usr/local/bin/register.sh
 COPY entrypoint.sh /usr/local/bin/entrypoint.sh
-RUN chmod +x /usr/local/bin/register.sh /usr/local/bin/entrypoint.sh
+RUN chmod +x /usr/local/bin/entrypoint.sh
 
 # Copy LiteFS configuration
 COPY litefs.yml /etc/litefs.yml

entrypoint.sh

@@ -1,8 +1,78 @@
 #!/bin/bash
+set -e
 
-# Start the registration loop in the background, redirecting output to stderr so we see it in Nomad logs
+# Configuration from environment
-/usr/local/bin/register.sh >&2 &
+SERVICE_NAME="navidrome"
+PORT=4533
+CONSUL_HTTP_ADDR="${CONSUL_URL:-http://localhost:8500}"
+NODE_IP="${ADVERTISE_IP}"
 
-# Start Navidrome
+# Tags for the Primary service (Traefik enabled)
-echo "Starting Navidrome..."
+PRIMARY_TAGS='["navidrome","web","traefik.enable=true","urlprefix-/navidrome","tools","traefik.http.routers.navidromelan.rule=Host(`navidrome.service.dc1.consul`)","traefik.http.routers.navidromewan.rule=Host(`m.fbleagh.duckdns.org`)","traefik.http.routers.navidromewan.middlewares=dex@consulcatalog","traefik.http.routers.navidromewan.tls=true"]'
-/app/navidrome
+
+NAVIDROME_PID=""
+SERVICE_ID="navidrome-${NODE_IP}-${SERVICE_NAME}"
+
+cleanup() {
+    echo "Caught signal, shutting down..."
+    if [ -n "$NAVIDROME_PID" ]; then
+        echo "Stopping Navidrome (PID: $NAVIDROME_PID)..."
+        kill -TERM "$NAVIDROME_PID"
+        wait "$NAVIDROME_PID" || true
+    fi
+    echo "Deregistering service ${SERVICE_ID} from Consul..."
+    curl -s -X PUT "${CONSUL_HTTP_ADDR}/v1/agent/service/deregister/${SERVICE_ID}" || true
+    exit 0
+}
+
+trap cleanup SIGTERM SIGINT
+
+echo "Starting leadership-aware entrypoint..."
+echo "Node IP: $NODE_IP"
+echo "Consul: $CONSUL_HTTP_ADDR"
+
+while true; do
+    # In LiteFS 0.5, .primary file exists ONLY on replicas.
+    if [ ! -f /data/.primary ]; then
+        # PRIMARY STATE
+        if [ -z "$NAVIDROME_PID" ] || ! kill -0 "$NAVIDROME_PID" 2>/dev/null; then
+            echo "Node is Primary. Initializing Navidrome..."
+            
+            # Register in Consul
+            echo "Registering as primary in Consul..."
+            curl -s -X PUT -d "{
+                \"ID\": \"${SERVICE_ID}\",
+                \"Name\": \"${SERVICE_NAME}\",
+                \"Tags\": ${PRIMARY_TAGS},
+                \"Address\": \"${NODE_IP}\",
+                \"Port\": ${PORT},
+                \"Check\": {
+                    \"HTTP\": \"http://${NODE_IP}:${PORT}/app\",
+                    \"Interval\": \"10s\",
+                    \"Timeout\": \"2s\"
+                }
+            }" "${CONSUL_HTTP_ADDR}/v1/agent/service/register"
+            
+            # Start Navidrome
+            /app/navidrome &
+            NAVIDROME_PID=$!
+            echo "Navidrome started with PID $NAVIDROME_PID"
+        fi
+    else
+        # REPLICA STATE
+        if [ -n "$NAVIDROME_PID" ] && kill -0 "$NAVIDROME_PID" 2>/dev/null; then
+            echo "Node transitioned to Replica. Stopping Navidrome..."
+            kill -TERM "$NAVIDROME_PID"
+            wait "$NAVIDROME_PID" || true
+            NAVIDROME_PID=""
+            
+            echo "Deregistering primary service from Consul..."
+            curl -s -X PUT "${CONSUL_HTTP_ADDR}/v1/agent/service/deregister/${SERVICE_ID}" || true
+        fi
+        
+        # We don't register anything for replicas in this version to keep it simple.
+        # But we stay alive so LiteFS keeps running.
+    fi
+    
+    sleep 5
+done

register.sh

@@ -1,73 +0,0 @@
-#!/bin/bash
-set -x
-
-# Configuration
-SERVICE_NAME="navidrome"
-REPLICA_SERVICE_NAME="replica-navidrome"
-PORT=4533
-CONSUL_HTTP_ADDR="${CONSUL_URL:-http://localhost:8500}"
-NODE_IP="${ADVERTISE_IP}"
-CHECK_INTERVAL="10s"
-
-# Tags for the Primary service (Traefik enabled)
-PRIMARY_TAGS='["navidrome","web","traefik.enable=true","urlprefix-/navidrome","tools","traefik.http.routers.navidromelan.rule=Host(`navidrome.service.dc1.consul`)","traefik.http.routers.navidromewan.rule=Host(`m.fbleagh.duckdns.org`)","traefik.http.routers.navidromewan.middlewares=dex@consulcatalog","traefik.http.routers.navidromewan.tls=true"]'
-
-# Tags for the Replica service
-REPLICA_TAGS='["navidrome-replica"]'
-
-register_service() {
-    local name=$1
-    local tags=$2
-    local id="navidrome-${NODE_IP}-${name}"
-    
-    echo "Registering as ${name} with ID ${id} at ${NODE_IP}:${PORT}..."
-    curl -v -X PUT -d "{
-        \"ID\": \"${id}\",
-        \"Name\": \"${name}\",
-        \"Tags\": ${tags},
-        \"Address\": \"${NODE_IP}\",
-        \"Port\": ${PORT},
-        \"Check\": {
-            \"HTTP\": \"http://${NODE_IP}:${PORT}/app\",
-            \"Interval\": \"${CHECK_INTERVAL}\",
-            \"Timeout\": \"2s\"
-        }
-    }" "${CONSUL_HTTP_ADDR}/v1/agent/service/register"
-}
-
-deregister_service() {
-    local name=$1
-    local id="navidrome-${NODE_IP}-${name}"
-    echo "Deregistering ${name} with ID ${id}..."
-    curl -v -X PUT "${CONSUL_HTTP_ADDR}/v1/agent/service/deregister/${id}"
-}
-
-echo "Starting Consul registration loop..."
-
-LAST_STATE="unknown"
-
-while true; do
-    # In LiteFS 0.5, .primary file exists ONLY on replicas.
-    # We check /data/.primary because /data is our mount point.
-    if [ ! -f /data/.primary ]; then
-        CURRENT_STATE="primary"
-    else
-        CURRENT_STATE="replica"
-    fi
-
-    echo "Current node state: ${CURRENT_STATE}"
-
-    if [ "$CURRENT_STATE" != "$LAST_STATE" ]; then
-        echo "State change detected: ${LAST_STATE} -> ${CURRENT_STATE}"
-        if [ "$CURRENT_STATE" == "primary" ]; then
-            deregister_service "$REPLICA_SERVICE_NAME"
-            register_service "$SERVICE_NAME" "$PRIMARY_TAGS"
-        else
-            deregister_service "$SERVICE_NAME"
-            register_service "$REPLICA_SERVICE_NAME" "$REPLICA_TAGS"
-        fi
-        LAST_STATE="$CURRENT_STATE"
-    fi
-    
-    sleep 15
-done