docker-cloud | project shows how to use Deployments | Continuous Deployment library

I'm trying to label each phaseJob in multiJob DSL plugin with specific node (in my case it's docker-cloud created with Docker Plugin).

Instead of labeling each phaseJob with specific node (docker-cloud) it labels every job with latest mentioned docker-cloud/node label.

I've tried the following example DSL configuration:

The answer is just to set the container to restart on crash. In development, I'm using nodemon which prevents the process from exiting on a crash - so I assumed that the process crashing didn't cause the container to exit. My mistake. The process crashing does in fact cause the container to exit when using node and not nodemon.

I’m trying to automate build on dockercloud to build a docker container for my front-end.

web.dockerfile

I'm trying to connect a service one in stack abc - abc_one to service two in stack zyx - zyx_two . These services share the network az-network .

I've read here that it's possible manually , but I need to do this with docker-compose.yml files for deploying to my swarm .

A simple version of of abc_one is

I'm learning about using Docker Compose to deploy applications in multiple containers, across multiple hosts. And I have come across two configuration files - stack file, and Compose file.

From the Cloud stack file YAML reference, it states a stack file is a file in YAML format that defines one or more services, similar to a docker-compose.yml file but with a few extensions.

And from this post, it states that stacks are very similar to docker-compose except they define services while docker-compose defines containers.

They look very similar, so I am wondering when I would use the stack file, and when to use the Compose file?

Background

I have a container, its running alot of stuff including a frontend that is exposed to other developers.

Users are going to be uploading some of their shell/python scripts unto my container to be run.

To keep the my container working, my plan is to send the script to a sibling container which will then run them and send me back the response. The user's scripts should be able to download external packages etc.

Then I want the sibling container to be "cleaned"

Question

Can I have that sibling container restart itself from its source image once it is done running the user's script? This way users can get a consistently clean container to run their scripts on.

Note

If I am completely barking up the wrong tree with this solution, please let me know. I am trying to get some weird functionalities going and could be approaching this from the wrong angle.

EDIT 1 (other approaches and why I don't think I like them)

Two alternatives that I have thought of is having the container with the frontend run containers on it. Or have the sibling container run docker containers on it. But these two solutions run into the difficulty of Docker-in-docker. The other solution may be to heighten my frontend container's permissions until it can make sibling containers on the fly for running scripts. But, I am worried that this may result in giving my frontend container unnecessarily high permissions.

EDIT 2 (all the documentation I have found on having a container restart itself)

I am aware of the documentation on autorestart, but I don't believe that this will clean the containers contents. For instance if a file was downloaded onto it.

I pulled an image I found on docker hub, I made some changes and it is working locally.

But now need to push it to docker hub. I used the following guide https://docs.docker.com/docker-cloud/builds/push-images/ and I was able to push the image on my repository.

The problem is that the changes I made are not in the image I just pushed. I think the reason is because I didn’t build (after making the changes) from the docker file. But given that the image I modified is not mine I don’t know where can I find(if I can ?) the DOCKERFILE, build the image and then push it.

Thank you

I am upgrading a docker-compose file from v1 to v3 and have a problem to link the containers together.

Before the docker-compose.yml was this:

I'm currently running Docker version 18.03.1-ce, build 9ee9f40 on multiple nodes. My setup is a nginx service and multiple java restful API services running in a wildfly cluster. For my API services I've configured a simple healthcheck to determine whether my API task is actually up: