From fifty-thousand feet, how are, for example, multiple MySQL containers handled?

In the context of my limited experience, simply running the container is insufficient, with additional configuration required. It just seems not quite so repeatable if, after running the container, it must be further tinkered with before any CRUD operations. Or is that just the process?


I think it depends on the context.

I feel your pain with regard to needing to tinker and tweak a MySQL/MariaDB instance. It's not extremely difficult to get an instance up and running and THEN populate the DB from, say, a mysqldump by attaching volumes to the MySQL instance either using a Dockerfile for the instance or by "manually" bash-ing into it and loading your tables, tweaking your network config, setting up complex user permissions, etc.

But it's usually a STRUGGLE to deploy a "ready to go" mysql instance from a docker-compose.yml in my initial experience using Docker Composer for a MySQL container in an app deployment, but it IS doable.

What kind of scale are you trying to achieve? Do you NEED Kubernetes or Docker Swarm? I think in the case of most apps, the answer is honestly no. You can almost always come up with a crafty workaround with ba(sh) scripts or RUN/CMD (Dockerfile) or command: (docker-compose).

Are you trying to support an automated deployment? or just get an app up and running? or trying to increase resilience? Increase capacity at scale?

  • those are the questions I should be asking. For the moment, I'd go with option #1: automated deployment. Sep 22 '20 at 22:18
  • 1
    docker-compose is likely what you want to use in conjunction with one or more Dockerfiles. You can build each individual container and configure your volumes and the docker network(s) in a myriad of (often ridiculously and confusing) ways by using different "versions" of docker-compose. I think the "latest" docker-compose version that seems to be in "wide spread use" is version: "3.7" which seems to have deprecated a TON of the k/v pairs used in docker-compose.yml files, like "depends_on:" and there are a few new (more "verbose") ways to granular config your docker volumes. Its a challenge..
    – K8sN0v1c3
    Sep 22 '20 at 22:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.