Docker 守护进程连接到套接字：2375

Question

我已将 Docker 1.12 版的 systemd 配置为侦听端口：2375。

[Unit]
Description=Docker Application Container Engine
Documentation=https://docs.docker.com
After=network.target

[Service]
Type=notify
ExecStart=/usr/bin/dockerd -H tcp://0.0.0.0:2375
ExecReload=/bin/kill -s HUP
MountFlags=slave
LimitNOFILE=1048576
LimitNPROC=1048576
LimitCORE=infinity
Delegate=yes
# kill only the docker process, not all processes in the cgroup
KillMode=process

[Install]
WantedBy=multi-user.target

结果：

*tcp6       0      0 :::2375                 :::*                    LISTEN*

但是，当我尝试在运行 Docker 的主机终端上使用“docker”命令时，我得到以下信息：

    [root@docker1-12 ~]# docker images
    Cannot connect to the Docker daemon. Is the docker daemon running on this host?

Even though the docker.service is actually running:

● docker.service - Docker Application Container Engine
   Loaded: loaded (/usr/lib/systemd/system/docker.service; enabled; vendor preset: disabled)
   Active: active (running) since Fri 2016-12-30 14:50:53 AEDT; 1min 32s ago
     Docs: https://docs.docker.com
 Main PID: 4976 (dockerd)
   Memory: 21.1M
   CGroup: /system.slice/docker.service
           ├─4976 /usr/bin/dockerd -H tcp://0.0.0.0:2375
           └─4982 docker-containerd -l unix:///var/run/docker/libcontainerd/docker-containerd.sock --shim docker-containerd-shim --metrics-interval=0 --start-timeout 2m --state-dir /var/run/docker/libcontainerd/containerd --runtime do...

Dec 30 14:50:52 docker1-12 dockerd[4976]: time="2016-12-30T14:50:52.083736426+11:00" level=info msg="[graphdriver] using prior storage driver \"devicemapper\""
Dec 30 14:50:52 docker1-12 dockerd[4976]: time="2016-12-30T14:50:52.091254467+11:00" level=info msg="Graph migration to content-addressability took 0.00 seconds"
Dec 30 14:50:52 docker1-12 dockerd[4976]: time="2016-12-30T14:50:52.095445562+11:00" level=info msg="Loading containers: start."
Dec 30 14:50:52 docker1-12 dockerd[4976]: time="2016-12-30T14:50:52.128643621+11:00" level=info msg="Firewalld running: true"
Dec 30 14:50:52 docker1-12 dockerd[4976]: time="2016-12-30T14:50:52.919797126+11:00" level=info msg="Default bridge (docker0) is assigned with an IP address 172.17.0.0/16. Daemon option --bip can be used to set a preferred IP address"
Dec 30 14:50:53 docker1-12 dockerd[4976]: time="2016-12-30T14:50:53.274835533+11:00" level=info msg="Loading containers: done."
Dec 30 14:50:53 docker1-12 dockerd[4976]: time="2016-12-30T14:50:53.275468481+11:00" level=info msg="Daemon has completed initialization"
Dec 30 14:50:53 docker1-12 dockerd[4976]: time="2016-12-30T14:50:53.275501250+11:00" level=info msg="Docker daemon" commit=7392c3b graphdriver=devicemapper version=1.12.5
Dec 30 14:50:53 docker1-12 dockerd[4976]: time="2016-12-30T14:50:53.285288956+11:00" level=info msg="API listen on [::]:2375"
Dec 30 14:50:53 docker1-12 systemd[1]: Started Docker Application Container Engine.

我该如何解决这个问题？即让 docker 监听端口：2375，同时能够在主机上本地运行 docker 命令？

注意 我找到了以下配置，作为一种解决方法。但是，这似乎不适用于 Docker 1.12 版：

Environment="DOCKER_OPTS=-H tcp://0.0.0.0:2375 --exec-opt native.cgroupdriver=cgroupfs"
ExecStart=/usr/bin/docker daemon -H fd://  \$DOCKER_OPTS

Answer 1

我能想到 2 个选项：

1.更改您的systemd单元文件：

ExecStart=/usr/bin/dockerd -H tcp://0.0.0.0:2375 -H unix:///var/run/docker.sock

这样，您将在 tcp 上侦听网络 API 调用，并在 unix sock 上侦听本地 docker cli。

2.尝试使用docker cli 连接时，请使用以下命令：

export DOCKER_HOST="tcp://0.0.0.0:2375"
docker ps

同样效果的一个衬里：

sudo docker -H tcp://0.0.0.0:2375 ps