docker container has been with us for many years. For software engineer which deal with legacy application, it can be intimidating to migrate to docker application. Because system stability, production investigation and quick fix and service stability surpasses every technology stack. At least that is for me, this cannot be compromise.
So where is docker container is best for? Where do we start docker container for? For many people, if you google, they build the application into docker image and deploy. Sure, that can be done but for me, as a start, I think for learning purposes and best use cases. Here, I would like share my experience and best use cases.
code compiling
Many of us do code compiling and this is especially true if you want to deploy a bunch of application, modify here and there, build the binary , get the final result and then move on. Installing software packages into container will not taint the host. Modify the configuration also need not worry about taint the host. What is even more merrier, you don't have to clean up after you done compiling. One can just delete the container and start over. Guaranteed reproducible end result. This is similar to chroot but container allow you to do further with different o/s.
application on legacy database
sometime when we migrate development server to the latest o/s but however it break the legacy application or its dependency, example database. In such a case, docker can come to the rescue, you can install legacy database in the docker container whilst the application still can access the database without any puzzle or not expected result. This is even more true when you are busy and focus on upgrading the development server, you don't have the time to upgrade the application or its database. When you are done develop development server, you can then upgrade the database to another version inside docker. I think docker in this case is a really good fit.
quick test for code
Many times when you write bash script, you want to quickly test if the script works fine. But you don't want to test run on your machine as it will change the underlying configuration/state. If you run a container, those code can be quickly test it and when you are done, you can exit the container and nothing harmful done to the host. In this case, you are totally sure your code is tested well and in the same time with less maintenance after tests.
That's it I want to share my opinions and use cases for docker container. If I learn new way of using docker, I will update this document. I hope you will find it useful for your use cases.