如何通过Docker Compose部署带有预准备数据的OpenSearch？

可以实现，官方未提及是因为这属于自定义部署场景，下面给两种适配你现有配置的可靠实现方式：

方法1：挂载预填充的数据目录（最直接） #

如果已经有导出的OpenSearch完整数据目录，直接替换默认空卷即可：

1. 先获取并处理预准备数据目录：
   从已运行的OpenSearch容器复制数据：

   docker cp opensearch-node1:/usr/share/opensearch-data ./opensearch-prepared-data
   

   确保目录权限匹配OpenSearch容器的运行用户（1000:1000）：

   chown -R 1000:1000 ./opensearch-prepared-data
   

2. 修改你的docker-compose.yml，将命名卷替换为本地预数据目录的绑定挂载：

   version: '3'
   services:
     opensearch-node1:
       image: opensearchproject/opensearch:latest
       container_name: opensearch-node1
       environment:
         - cluster.name=opensearch-cluster
         - node.name=opensearch-node1
         - discovery.type=single-node
         - bootstrap.memory_lock=true
         - "OPENSEARCH_JAVA_OPTS=-Xms1024m -Xmx1024m"
       ulimits:
         memlock:
           soft: -1
           hard: -1
         nofile:
           soft: 65536
           hard: 65536
       volumes:
         # 替换为本地预准备的数据目录路径
         - ./opensearch-prepared-data:/usr/share/opensearch/data
       ports:
         - 9200:9200
         - 9600:9600
     opensearch-dashboards:
       image: opensearchproject/opensearch-dashboards:latest
       container_name: opensearch-dashboards
       ports:
         - 5601:5601
       expose:
         - "5601"
       environment:
         OPENSEARCH_HOSTS: '["https://opensearch-node1:9200"]'
   

   启动后，OpenSearch会直接加载预存数据，无需额外导入操作。

方法2：添加初始化脚本自动导入数据（适合动态数据文件） #

如果只有批量导入的JSON数据文件，可新增一个初始化服务，等OpenSearch启动就绪后自动执行导入：

1. 准备符合OpenSearch批量API格式的数据文件（比如./init-data/bulk-data.json）：

   {"index": {"_index": "my-index", "_id": "1"}}
   {"title": "Test Document 1", "content": "This is pre-prepared data"}
   {"index": {"_index": "my-index", "_id": "2"}}
   {"title": "Test Document 2", "content": "More pre-prepared data"}
   

2. 修改docker-compose.yml，新增初始化服务：

   version: '3'
   services:
     opensearch-node1:
       image: opensearchproject/opensearch:latest
       container_name: opensearch-node1
       environment:
         - cluster.name=opensearch-cluster
         - node.name=opensearch-node1
         - discovery.type=single-node
         - bootstrap.memory_lock=true
         - "OPENSEARCH_JAVA_OPTS=-Xms1024m -Xmx1024m"
       ulimits:
         memlock:
           soft: -1
           hard: -1
         nofile:
           soft: 65536
           hard: 65536
       volumes:
         - opensearch-data1:/usr/share/opensearch/data
       ports:
         - 9200:9200
         - 9600:9600
       healthcheck:
         # 健康检查确保OpenSearch就绪
         test: ["CMD-SHELL", "curl -s --insecure https://localhost:9200/_cluster/health | grep -q '\"status\":\"green\"'"]
         interval: 10s
         timeout: 10s
         retries: 5
     opensearch-dashboards:
       image: opensearchproject/opensearch-dashboards:latest
       container_name: opensearch-dashboards
       ports:
         - 5601:5601
       expose:
         - "5601"
       environment:
         OPENSEARCH_HOSTS: '["https://opensearch-node1:9200"]'
       depends_on:
         opensearch-node1:
           condition: service_healthy
     # 新增初始化数据服务
     opensearch-init:
       image: curlimages/curl:latest
       container_name: opensearch-init
       volumes:
         - ./init-data:/init-data
       depends_on:
         opensearch-node1:
           condition: service_healthy
       command: >
         sh -c "curl -s --insecure --user admin:admin -X POST https://opensearch-node1:9200/_bulk -H 'Content-Type: application/json' --data-binary @/init-data/bulk-data.json"
       # 执行完成后自动退出
       restart: "no"
   volumes:
     opensearch-data1:
   

   启动后，初始化服务会在OpenSearch就绪后自动导入数据，完成后容器自动停止。

注意事项 #

• 方法1中，预数据目录必须与OpenSearch版本兼容，避免版本不匹配导致数据损坏。
• 方法2中，默认OpenSearch启用了HTTPS和账号认证，curl命令需带上--insecure和--user admin:admin（若修改了默认密码，需对应调整）。

内容的提问来源于stack exchange，提问作者Руслан Кузнецов