Hi,
I am pretty new to world of docker and even more to the rancher farm so maybe my question will be dummy.
I have deploy my first rancher2 cluster and an additional node. Now I want to deploy a docker image from http://hub.docker.com/. So I have follow the steps in https://rancher.com/docs/rancher/v2.0/en/quick-start-guide/#deploy-a-workload but I am confused at that step… I was using previously portainer to deploy my docker image and when I configure a new image, it extract the information about variables, ports, etc. Of course I can configure some stuff but most of the configuration work is already done by I guess extracting the information from the docker image. With rancher2 and the workload, I am pretty lost… do I need to figure out the variables and the ports by myself? Or do I have missed a step somewhere?
Thanks for your help!
Yeah, you have to read the Docker Hub documentation to find the ports and environment variables used by the image.
We tried this a long time ago (before 1.0) and it doesn’t work very well. Talking to the registry is slow (especially trying to do autocomplete/search for image names). The server container doesn’t necessarily have access to the registry so the communication has to be proxied to the node that’s going to pull the image, but you don’t know which one that is yet because it’s not scheduled yet. The image is not necessarily in a registry at all but might be already just on the node (but again, you don’t know which node to look at without scheduling).
…And at the end you find out that many people’s images don’t actually bother to populate port/env metadata. Or if they do, it’s often partially complete/partially missing or wrong, because nothing actually depends on it being correct.
1 Like