MySQL - populate the pipelines database with your tables and schema

Question

When it comes to initializing mysql database (on a service container), the documentation refers to the official docker hub MySQL page. Normally one would start a container and mount host volume with init .sql files that would be picked up by docker container and ran from mounted volume /docker-entrypoint-initdb.d.

I hit the wall trying to figure out how to this in the context of the bitbucket-pipelines.yml.

Any help would be greatly appreciated!

- Michael

Misha Shirman · Answer

Thank you Max!

I though i would need to go this route, but wanted to keep the build times to a minimum. Thinking maybe we will maintain our custom db container separately and use it for the test step as a service. So that we don't have to build/push/pull a service container while building the app.

MB_4f5oph · Answer

In your pipeline, you could do the following;

* In your pipeline, create a Dockerfile from the MySQL base image, and copy your .sql or .sh scripts to the entrypoint
* Build and Push the image to a registry of your choice ( ie private, local, google container registry etc )
* Use service definitions in pipelines to use the image you pushed containing your .sql
* Parse environment variables to the service definition such as MYSQL_DATABASE, MY_SQLPASSWORD etc

Some more details on definitions can be found here.