docker run 成功， docker compose up 失败？

[wganbleuthall]

docker run -it -p 9201:9200 --network elastic --name es2 elasticsearch:8.7.1没有任何问题，几秒之后就会出现 ✅ Elasticsearch security features have been automatically configured! ✅ Authentication is enabled and cluster connections are encrypted.

---

但是 docker-compose.yml

version: "3.9"
services:
  logstash:
    image: logstash:8.7.1
    container_name: logstash
    networks:
      - elastic
    ports:
      - "5044:5044"
    volumes:
      - ./elk/pipeline:/usr/share/logstash/pipeline:rw,Z
    restart: unless-stopped
    depends_on:
      - kibana
  kibana:
    image: kibana:8.7.1
    container_name: kibana
    networks:
      - elastic
    ports:
      - "5601:5601"
    restart: always
  elasticsearch:
    image: elasticsearch:8.7.1
    container_name: elasticsearch
    deploy:
      resources:
        reservations:
          memory: 4GB
    networks:
      - elastic
    ports:
      - "9200:9200"
    restart: no


networks:
  elastic:
    name: elastic
    external: true

这样子启动的 elasticsearch 就会无限打印

{"@timestamp":"2023-06-02T12:25:51.290Z", "log.level": "WARN", "message":"received plaintext http traffic on an https channel, closing connection Netty4HttpChannel{localAddress=/172.19.0.3:9200, remoteAddress=/172.19.0.2:42386}", "ecs.version": "1.2.0","service.name":"ES_ECS","event.dataset":"elasticsearch.server","process.thread.name":"elasticsearch[bb66f54e0118][transport_worker][T#2]","log.logger":"org.elasticsearch.http.netty4.Netty4HttpServerTransport","elasticsearch.cluster.uuid":"7sT8niyaQeqs2oZAn8TLLg","elasticsearch.node.id":"jg-nmYbcQZmdM5Jfy_fMfw","elasticsearch.node.name":"bb66f54e0118","elasticsearch.cluster.name":"docker-cluster"}

然后几分钟后就ERROR: Elasticsearch exited unexpectedly,为什么，我无法理解了，还没有任何的 error 日志 它要是显示启动成功的那两个绿勾勾再无限打印 warning 也就算了，可是压根启动不成功

---

一开始我以为是 docker compose 启动过程中 kibana 不停给 elasticsearch 发 http 导致启动失败，但是 depends-on 完全不起作用，甚至我将 docker compose yml 删了只剩 elasticsearch 一个服务还是不行 但是 docker run ...就没有那串日志，所以难道 docker compose 在启动的时候，docker engine 不停给 elasticsearch 发什么 http 吗，healthcheck?? 这我该咋办啊

[Yourshell]

有没有可能是内存给少了

[hingle]

docker run 端口是 9201 ，docker compose 端口是 9200 ，端口被占用了？

[wganbleuthall]

@Yourshell 不应该啊 给了 4g

[wganbleuthall]

@hingle 两个容器

[caomingjun]

我感觉不像 healthcheck ，healthcheck 一般是从容器内部发起的
其实日志输出里有 remoteAddress=/172.19.0.2:42386 ，你搞清楚这个 ip 和端口号上跑的啥玩意不就知道了

[w07128597]

我遇到过，默认内存太低了，启动成功后立马就挂了，增加 es 的堆内存就好了，在环境变量中设置

[chotow]

docker inspect elasticsearch 可以查看退出原因，我试了下，是因为 OOM