openmediavault-docker-gui Testing

Quote from ryecoaaron

This is a warning not an error. It is fine. The plugin just needs to be updated for the latest OMV changes.

Ah, well I'd give it a go but I was only testing if I wanted the plugin. I'll stick with cli.

Quote from Pheoxy

You need to stop container and pull again is the easiest way.

Code

docker stop container
docker pull image/image

Thanks, that almost did it. After pulling the new version the container kept using the old version.
I modified the container (changing nothing, just clicking modify then save) and then it started using the latest version.

An easier way is to setup a systemd starup script with:

docker rm container
docker pull image/image
docker run --name=container

Code parts in it. I can post an example when I'm not on my phone.

Help !Cannot install docker.
OMV3

g2_ufo: You apparently need to enable both docker and docker testing repo in OMV-extras, then it should work

Many thanks!

I have a brand new OMV installation (2.2.13).
When I install docker and activate it, I get this error:

Fehler
Error: "couldn't connect to host" - Code: 7
Fehler #0:
exception 'OMVModuleDockerException' with message 'Error: "couldn't connect to host" - Code: 7' in /usr/share/omvdocker/Utils.php:63
Stack trace:
#0 /usr/share/omvdocker/Utils.php(180): OMVModuleDockerUtil::doApiCall('http://localhos...')
#1 /usr/share/openmediavault/engined/rpc/docker.inc(354): OMVModuleDockerUtil::getContainers('42005')
#2 [internal function]: OMVRpcServiceDocker->getContainers(Array, Array)
#3 /usr/share/php/openmediavault/rpcservice.inc(125): call_user_func_array(Array, Array)
#4 /usr/share/php/openmediavault/rpc.inc(79): OMVRpcServiceAbstract->callMethod('getContainers', Array, Array)
#5 /usr/sbin/omv-engined(500): OMVRpc::exec('Docker', 'getContainers', Array, Array, 1)
#6 {main}

I even tried to change the port:

Fehler #0:
exception 'OMVModuleDockerException' with message 'Error: "couldn't connect to host" - Code: 7' in /usr/share/omvdocker/Utils.php:63
Stack trace:
#0 /usr/share/omvdocker/Utils.php(125): OMVModuleDockerUtil::doApiCall('http://localhos...')
#1 /usr/share/openmediavault/engined/rpc/docker.inc(145): OMVModuleDockerUtil::getImages('12345', true)
#2 [internal function]: OMVRpcServiceDocker->getImages(Array, Array)
#3 /usr/share/php/openmediavault/rpcservice.inc(125): call_user_func_array(Array, Array)
#4 /usr/share/php/openmediavault/rpc.inc(79): OMVRpcServiceAbstract->callMethod('getImages', Array, Array)
#5 /usr/sbin/omv-engined(500): OMVRpc::exec('Docker', 'getImages', Array, Array, 1)
#6 {main}

New omv ystem. Docker is the first plugin I installed. It can reproduce the problem with another new installation.
It's a VM on Hyper-V Server 2016.

It's not possible to start docker.

service start docker
-> starting docker: docker
docker ps -a
-> Cannot connect to the Docker daemon. Is the docker daemon running on this host?

Edit: Docker.log says, that my Kernel 3.2.0-4amd64 is not supported.
i will try again with backports.

Edit2: Works!

Hello,

just got this error randomly. I had to restart docker service from commandline. I can click OK and everything works as intended, just wanted to inform

Error

Object of class stdClass could not be converted to string
OKShow details
Error

Object of class stdClass could not be converted to string
Error #0:exception 'ErrorException' with message 'Object of class stdClass could not be converted to string' in /usr/share/openmediavault/engined/rpc/docker.inc:1461Stack trace:#0 /usr/share/openmediavault/engined/rpc/docker.inc(1461): errorHandler(4096, 'Object of class...', '/usr/share/open...', 1461, Array)#1 [internal function]: OMVRpcServiceDocker->getDockerRepo(Array, Array)#2 /usr/share/php/openmediavault/rpc/serviceabstract.inc(124): call_user_func_array(Array, Array)#3 /usr/share/php/openmediavault/rpc/rpc.inc(84): OMV\Rpc\ServiceAbstract->callMethod('getDockerRepo', Array, Array)#4 /usr/sbin/omv-engined(516): OMV\Rpc\Rpc::call('Docker', 'getDockerRepo', Array, Array, 1)#5 {main}

I've latest OMV3 and docker-testing plugin from OMV-Extras installed. When pulling simple ubuntu:trusty image and running this container, there is an error. GUI prints only

Restarting (0) Less than a second ago

all the time. But running container on OMV shell with

docker run -t -i ubuntu:trusty

all is fine. Syslog contains

Dec 14 20:13:01 omv-dev docker[1090]: time="2016-12-14T20:13:01.999562243+01:00" level=error msg="containerd: notify OOM events" error="cgroup path for memory not found"

What's the reason for this? And may this error in syslog causes Docker GUI to fail starting this container?

Quote from nicjo814

The plug-in runs the container demonized which I think it's not possible with the container you are trying to run. The plug-in is mainly intended for running docker apps as "services".

You are right. I'm new to Docker, but I'm learning a lot now. Cgroup errors in syslog disappeared with help of this manual.
And yes, you are right. Running container on shell with "docker run ubuntu:trusty" exits immediately, only "docker run -t -i ubuntu:trusty" keeps a shell opened. Using "/sbin/init" as CMD instead of default "/bin/bash" keeps the container running. But why does "/bin/bash" does not work? Is it because it will open a shell on a TTY that does not exist?

Ok, I think I understand the core concept and what to do to get a container running in GUI.

@nicjo814 One thing I am missing is: when modifying an existing container, one cannot modify the value of CMD. Some 3rd party containers have a default CMD which does not work with daemon mode (see my last post and "/bin/bash"). But to prevent rebuilding this image a simple solution could be to modify the CMD value as one can do with "docker run [...]". Are you going to implement this?

The popup in GUI provides no field where I can type in a different CMD. The CMD in docker command has to be the last option on CLI. So it has to be "docker run -d [...] [CMD]"
The field "Extra args" in GUI prepends the input value to the container name, like

Failed to execute command 'export PATH=/bin:/sbin:/usr/bin:/usr/sbin:/usr/local/bin:/usr/local/sbin; export LANG=C; docker run -d --restart=always -v /etc/localtime:/etc/localtime:ro --net=bridge -h "40af611f3e5a" -e PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin" --name="elegant_varahamihira" /sbin/init "ubuntu:trusty" 2>&1' with exit code '125': docker: Error parsing reference: "/sbin/init" is not a valid repository/tag. See 'docker run --help'.

As you can see it puts "/sbin/init" before "ubuntu:trusty", so I think there has to be a special input field in GUI for managing this.

Ok. I get it to work in detached mode with

docker run -d ubuntu:trusty /sbin/init
docker run -d ubuntu:trusty ping localhost

Normally there should be no other app in container than the one specified by CMD. *If* a CMD is specified, there can also be containers without any declaration of CMD. Also, when using ENTRYPOINT instead of CMD one could pass CLI arguments to executed script inside container and that's the same syntax as additional arguments for entrypoint have to be passed after the container name.