fix(docker): 健康检查加固 + kafka/zk 僵尸进程根治 + 移除死依赖

docker-compose.yml

@@ -49,6 +49,9 @@ services:
     image: confluentinc/cp-zookeeper:7.5.0
     container_name: mes-zookeeper
     restart: always
+    # Confluent entrypoint 不 exec 且健康检查脚本频繁 fork，僵尸子进程无人回收
+    # (服务器实测 137 个)；init: true 注入 tini 作为 PID 1 统一回收
+    init: true
     ports:
       - "2181:2181"
     environment:
@@ -71,6 +74,8 @@ services:
     image: confluentinc/cp-kafka:7.5.0
     container_name: mes-kafka
     restart: always
+    # 同上：僵尸进程累积 (服务器实测 1072 个)，tini 作为 PID 1 回收
+    init: true
     depends_on:
       zookeeper:
         condition: service_healthy

mes-auth/Dockerfile

@@ -32,7 +32,7 @@ chmod +x /app/start.sh
 
 EXPOSE 8081
 
-HEALTHCHECK --interval=30s --timeout=3s --retries=3 \
+HEALTHCHECK --interval=60s --timeout=10s --retries=3 --start-period=90s \
   CMD curl -sf http://localhost:8081/actuator/health || exit 1
 
 CMD ["/app/start.sh"]

mes-dashboard/Dockerfile

@@ -32,7 +32,7 @@ chmod +x /app/start.sh
 
 EXPOSE 8085
 
-HEALTHCHECK --interval=30s --timeout=3s --retries=3 \
+HEALTHCHECK --interval=60s --timeout=10s --retries=3 --start-period=90s \
   CMD curl -sf http://localhost:8085/actuator/health || exit 1
 
 CMD ["/app/start.sh"]

mes-gateway/Dockerfile

@@ -32,7 +32,7 @@ chmod +x /app/start.sh
 
 EXPOSE 9090
 
-HEALTHCHECK --interval=30s --timeout=3s --retries=3 \
+HEALTHCHECK --interval=60s --timeout=10s --retries=3 --start-period=90s \
   CMD curl -sf http://localhost:9090/actuator/health || exit 1
 
 CMD ["/app/start.sh"]

mes-process/Dockerfile

@@ -32,7 +32,7 @@ chmod +x /app/start.sh
 
 EXPOSE 8083
 
-HEALTHCHECK --interval=30s --timeout=3s --retries=3 \
+HEALTHCHECK --interval=60s --timeout=10s --retries=3 --start-period=90s \
   CMD curl -sf http://localhost:8083/actuator/health || exit 1
 
 CMD ["/app/start.sh"]

mes-quality/Dockerfile

@@ -32,7 +32,7 @@ chmod +x /app/start.sh
 
 EXPOSE 8084
 
-HEALTHCHECK --interval=30s --timeout=3s --retries=3 \
+HEALTHCHECK --interval=60s --timeout=10s --retries=3 --start-period=90s \
   CMD curl -sf http://localhost:8084/actuator/health || exit 1
 
 CMD ["/app/start.sh"]

mes-workorder/Dockerfile

@@ -32,7 +32,7 @@ chmod +x /app/start.sh
 
 EXPOSE 8082
 
-HEALTHCHECK --interval=30s --timeout=3s --retries=3 \
+HEALTHCHECK --interval=60s --timeout=10s --retries=3 --start-period=90s \
   CMD curl -sf http://localhost:8082/actuator/health || exit 1
 
 CMD ["/app/start.sh"]

pom.xml

@@ -40,7 +40,6 @@
         <jjwt.version>0.12.5</jjwt.version>
         <fastjson2.version>2.0.47</fastjson2.version>
         <influxdb-client.version>1.35.0</influxdb-client.version>
-        <elasticsearch-java.version>8.10.0</elasticsearch-java.version>
         <kafka-clients.version>4.3.1</kafka-clients.version>
         <mqttv5-client.version>5.1.4</mqttv5-client.version>
         <seata.version>2.0.0</seata.version>
@@ -122,11 +121,6 @@
                 <artifactId>influxdb-client-java</artifactId>
                 <version>${influxdb-client.version}</version>
             </dependency>
-            <dependency>
-                <groupId>co.elastic.clients</groupId>
-                <artifactId>elasticsearch-java</artifactId>
-                <version>${elasticsearch-java.version}</version>
-            </dependency>
             <dependency>
                 <groupId>org.apache.kafka</groupId>
                 <artifactId>kafka-clients</artifactId>