[Container-tools] outside access to Docker daemon in CDK

[Keith Babo]

Howdy,

I’ve been playing around with the CDK a bit in preparation for the xPaaS release of Fuse for OpenShift 3.1.  Things are going quite well, but I’ve hit one small snag.  It appears as though the Docker daemon is meant to be exposed outside the vagrant image based on ‘vagrant adbinfo’:

export DOCKER_HOST=tcp://10.1.2.2:2376
export DOCKER_CERT_PATH=/Users/kbabo/Downloads/cdk/components/rhel/rhel-ose/.vagrant/machines/default/virtualbox/.docker
export DOCKER_TLS_VERIFY=1
export DOCKER_MACHINE_NAME=52c3900

Unfortunately, I can’t connect with a docker client to 10.1.2.2:2376 outside the image.  Nothing at that port and looking at /etc/sysconfig/docker and the docker process itself, there does not appear to be a -H host directive for a TCP binding.  I can fix this by updating the OPTIONS var in /etc/sysconfig/docker to include -H unix:///var/run/docker.sock -H tcp://0.0.0.0:2375 , but I’m guessing this is not the intended path.

Aside from the above, the CDK fits our initial use case like a glove.  I want to make sure we document this for our users, so any insight you can share on what I might be messing up and/or the appropriateness of my workaround above would be appreciated.

BTW, I’m using the Beta 3 build available from the CSP.

thanks,
keith

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/container-tools/attachments/20151210/598b886b/attachment.htm>