docker elasticsearch容器不转发端口(macOs)

如果您在OSX上运行docker,请注意主机实际上是初始化boot2docker时安装的VirtualBox实例.所以在这种情况下,而不是使用:

curl http://localhost:9200

找到虚拟盒实例的IP,我将其称为VM_IP:

boot2docker ip

然后尝试:

curl http://<VM_IP>:9200

在docker容器中运行Elasticsearch时,我遇到了端口转发问题.我通过手动指定绑定Elasticsearch服务器的接口解决了这个问题docker run --rm -p 9200:9200 -p 9300:9300 --name=es elasticsearch:latest -Des.network.host=0.0.0.0.

绑定部分是-Des.network.host=0.0.0.0最后的部分.我在https://mad.is/2016/09/running-elasticsearch-in-docker-container/写了一篇博文,详细说明了这一点.

我测试了你的Dockerfile,但它确实有效.

FROM dockerfile/java
RUN \
  cd /tmp && \
  wget https://download.elasticsearch.org/elasticsearch/elasticsearch/elasticsearch-1.3.2.tar.gz && \
  tar xvzf elasticsearch-1.3.2.tar.gz && \
  rm -f elasticsearch-1.3.2.tar.gz && \
  mv /tmp/elasticsearch-1.3.2 /elasticsearch
# Define mountable directories.                                                                                    
VOLUME ["/data"]
# Define default command.                                                                                          
CMD ["/elasticsearch/bin/elasticsearch"]
EXPOSE 9200
EXPOSE 9300

我尝试构建这个Dockerfile并运行它.

$ docker build -t 25312935 .
$ docker run -t -p 9200:9200 -p 9300:9300 --rm 25312935
[2014-08-15 04:41:08,349][INFO ][node                     ] [Black Crow] version[1.3.2], pid[1], build[dee175d/2014-08-13T14:29:30Z]
[2014-08-15 04:41:08,349][INFO ][node                     ] [Black Crow] initializing ...
[2014-08-15 04:41:08,353][INFO ][plugins                  ] [Black Crow] loaded [], sites []
[2014-08-15 04:41:10,444][INFO ][node                     ] [Black Crow] initialized
[2014-08-15 04:41:10,444][INFO ][node                     ] [Black Crow] starting ...
[2014-08-15 04:41:10,547][INFO ][transport                ] [Black Crow] bound_address {inet[/0:0:0:0:0:0:0:0:9300]}, publish_address {inet[/172.17.0.72:9300]}
[2014-08-15 04:41:10,560][INFO ][discovery                ] [Black Crow] elasticsearch/0mpczYoYSZCiAmbkxcsfpg
[2014-08-15 04:41:13,601][INFO ][cluster.service          ] [Black Crow] new_master [Black Crow][0mpczYoYSZCiAmbkxcsfpg][eeb3396b1ecc][inet[/172.17.0.72:9300]], reason: zen-disco-join (elected_as_master)
[2014-08-15 04:41:13,615][INFO ][http                     ] [Black Crow] bound_address {inet[/0:0:0:0:0:0:0:0:9200]}, publish_address {inet[/172.17.0.72:9200]}
[2014-08-15 04:41:13,615][INFO ][node                     ] [Black Crow] started
[2014-08-15 04:41:13,634][INFO ][gateway                  ] [Black Crow] recovered [0] indices into cluster_state

如下所示,request 127.0.0.1:9200返回json响应.

$ curl 127.0.0.1:9200
{
  "status" : 200,
  "name" : "Black Crow",
  "version" : {
"number" : "1.3.2",
  },
  "tagline" : "You Know, for Search"
}

检查你的-p选择.这意味着将容器的端口公开给主机.如果您没有明确写入主机端口,则docker会像下面一样分配随机端口.

$ docker run -t -p 9200 -p 9300 --rm 25312935
$ docker ps -l
CONTAINER ID        IMAGE                COMMAND                CREATED             STATUS              PORTS                                              NAMES
1aa4c2c84d04        25312935:latest      /elasticsearch/bin/e   15 seconds ago      Up 15 seconds       0.0.0.0:49153->9200/tcp, 0.0.0.0:49154->9300/tcp   sad_shockley

0.0.0.0:49153->9200/tcp 意味着您可以通过主机的49153端口访问容器的9200端口.

$ curl 127.0.0.1:49153
{
  "status" : 200,
  "name" : "Golem",
  "version" : {
"number" : "1.3.2",
  },
  "tagline" : "You Know, for Search"
}

所以如果你想使用主机s 9200 port, explicitly write the host port like-p 9200:9200 or-p 0.0.0.0:9200:9200`

$ docker ps -l
CONTAINER ID        IMAGE                COMMAND                CREATED             STATUS              PORTS                                            NAMES
eeb3396b1ecc        25312935:latest      /elasticsearch/bin/e   59 seconds ago      Up 58 seconds       0.0.0.0:9200->9200/tcp, 0.0.0.0:9300->9300/tcp   high_elion

如果仍然无效,请尝试--net=host选项.您可以使用此选项在容器内使用主机网络堆栈.

$ docker run -t --net=host --rm 25312935
$ curl 127.0.0.1:9200
{
  "status" : 200,
  "name" : "Black Crow",
  "version" : {
"number" : "1.3.2",
  },
  "tagline" : "You Know, for Search"
}

如果两者都不起作用,我认为您需要检查其他网络配置.