Difference between development and production Docker?

[hirokiterashima]

Question from @HsienLu

I am a developer from National Taiwan University of Science and Technology. Thank you for providing such a wonderful open platform. We are planning to use this platform as one of the tools for our research project.

However, I have two questions that I would like to consult with you:

In your README, there are two versions: a development version and a production version. Could you kindly explain the difference between the two and why this distinction is necessary?
If I use the development version for local development, how should I update to the production version after completing the development process?
Thank you for your time and assistance.

Best regards,
HsienLu

[hirokiterashima]

Hi HsienLu,

Welcome to the forum, and thanks for using WISE. We're excited to hear that you are planning to use the platform for your research project.

Differences between the two versions of Docker

We use the WISE-Docker-Dev repo for developing WISE locally, as you know. The Docker containers that are referenced in WISE-Docker-Dev/docker-compose.yml use the wise-client-dev and wise-api-dev Docker images, which contain tools to watch for changes and compile the codebase.

WISE-Docker-Server is used to deploy pre-built WISE Docker images on a server that is running Docker. When we make a release for WISE-Client or WISE-API on GitHub, we build and publish a Docker image (using GitHub actions, for example here: https://github.com/WISE-Community/WISE-Client/actions/workflows/publish-docker-image-prod.yml). This image contains the production-ready artifacts (static files for WISE-Client, and a WAR file for WISE-API), and does not contain devtools to watch/compile code.

Deploying your own WISE instance

As described on our README, there are two options for deploying WISE: 1) using Docker, and 2) not using Docker.

Option 1: using Docker

This option requires you to build your own Docker image(s), publish them (on somewhere like https://hub.docker.com) and reference them in WISE-Docker-Server/docker-compose.yml

If you set up your GitHub repo to build Docker images on release (e.g., using GitHub actions) and configure DockerHub credentials so that the images can be published there, then you just need to edit the docker-compose.yml file and you should be good to go.

Option 2: not using Docker

You'd be following the instructions on this page: https://github.com/WISE-Community/.github/wiki/How-to-set-up-WISE-on-your-production-server

Basically, this option requires you to

1. Set up the server (install Nginx, Tomcat, Mysql, Redis etc)
2. Build your project locally (static files for WISE-Client, and a WAR file for WISE-API) and put them in the right place on the server

I hope this helps. Let me know if you have any questions!

Hiroki

[HsienLu]

Hi Hiroki,

Thank you for your clear explanation. We are planning to use Docker for development, as it seems to be both quick and convenient.

From my understanding, for local development, I would package the modified wise-client and wise-api folders into Docker images and then reference these images when deploying to the production version. This would handle both deployment and updates.

Is my understanding correct, or is there anything I might have misunderstood?

Thank you again for your assistance!

Best regards,
HsienLu

[hirokiterashima]

Yes, your understanding is correct. The easiest is to clone the repos and enable the GitHub actions to build and push the Docker images to DockerHub when new code is committed to the repos. This is what we do.

Alternatively, you can manually build the Docker images locally and import them to your server.

Let me know if you have any more questions.

[HsienLu]

Thank you for your response. I will try this workflow, and I will keep you updated on the results. Thank you again!

[HsienLu]

I'm sharing two issues I encountered and their solutions while setting up a local development environment using Docker Compose on a "Windows 10."

---

Issue 1 sh: /bin/bash^M: bad interpreter: No such file or directory

• Problem Description: When starting the project locally on Windows, the following error message appears:

  /bin/bash^M: bad interpreter: No such file or directory

• My Solution: Modify the command field in the docker-compose.yml file under the wise-api service. Change the original command:

  command: >
    bash -c '/app/create-application-dockerdev-properties-if-necessary.sh &&
    mvn spring-boot:run -Dspring-boot.run.jvmArguments="-Xdebug -Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=*:5005" -Dspring-boot.run.profiles=dockerdev'

  To:

  command: >
    bash -c 'sed -i "s/\r$//" /app/create-application-dockerdev-properties-if-necessary.sh &&
    /app/create-application-dockerdev-properties-if-necessary.sh &&
    mvn spring-boot:run -Dspring-boot.run.jvmArguments="-Xdebug -Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=*:5005" -Dspring-boot.run.profiles=dockerdev'

Issue 2 "PWD" variable not set

• Problem Description: When starting the project locally on Windows, the PWD environment variable is not set, causing the following error message to appear.This prevents localhost:81 from running properly.

  level=warning msg="The \"PWD\" variable is not set. Defaulting to a blank string.

• My Solution: Before running docker-compose, manually set the PWD variable:

  $env:PWD = Get-Location