Trellis/Ansible + SQL Scripts + Rollback

[andrewfrankel]

Searched for a while with no avail, but please point me in the right direction if I missed it…

Wondering the best practices for including SQL scripts in the trellis/ansible deployment process (possibly with relation to Wordpress’ dbDelta for schema updates) and additionally how to revert those database changes on rollback.

The SQL scripts might include schema and/or data changes. Ideally a specific script could be configured to execute on the next/specific deployment but never again.

Thanks in advance!

[kalenjohnson]

Trellis doesn’t handle database updates by default.

What you’re describing sounds like database migrations and seeding, something that is not included with WordPress, and not something I’ve ever actually seen happen with it. But holy crap could WP use it, however it’s a very developer centric feature so nothing that will ever be added into WP core.

Many people once they get into a multiple developer project, have only a couple options that I’ve seen used successfully. I think that generally, developers need to be in communication about who’s doing what with the database.

There were some interesting suggestions here: https://deliciousbrains.com/avoiding-wordpress-database-merging/

This can help with keeping site settings in version control: https://github.com/danielbachhuber/dictator

This is an expensive plugin but really the only one I’ve seen that is able to do a true database comparison and merge: https://crowdfavorite.com/ramp/

This has also been discussed quite a few times previously:

Keep database synced between computers in Trellis/Bedrock/Sage setup

Hi there, I'm trying out the new Trellis/Bedrock/Sage setup for the first time and I'm running into some questions; My previous workflow for development websites was with MAMP Pro where I synced my database and htdocs folder with dropbox so I could work from both my computer at work and at home on the same project. Managing my databases was done with phpMyAdmin so I could export my database, find and replace path and domain names and import to my production environment. Now with Trellis I …

Syncing databases between mulitple developers and server

I have two developers working on local copies of a WP build and deploying to a staging server via capistrano. When we run wpcli:db:push, rather than syncing the database on the server, it completely overrides it. For example, Dev A makes adds custom fields on his machine and pushes DB to server. Dev B makes changes to a page and pushes to server. Only the Dev B changes remain. Is this expected behavior? Is there a solution/workflow that accommodates a sync?

Best practices for syncing the development database between multiple developers/machines with Bedrock Ansible?

Hey all, I've just recently started using bedrock with Ansible, having used composer for a few previous projects. The benefits of using Ansible seem quite clear to me, having just spent 5 minutes provisioning a box which would normally take a lot longer. I'm having difficulty figuring out how I'd go about syncing the WordPress database with multiple development machines, or at least being able to use the latest version of the database. Previously, I've tried having a master database, that …

[andrewfrankel]

@kalenjohnson, thanks for the quick reply!

More so than migrations/seeding/syncing, we need to normalize/alter old data to work with newly optimized functionality (though maybe those can be handled via a migration process). Conceptually, I feel most comfortable being able to run straight queries for those types of tasks. I.E. on deployment run a sets of sql scripts.

The big part I didn’t see covered is rollback - after making all these changes to the database, how would one easily revert to a prior state? I figured something along the lines of making a backup of the DB before deploy, which could be imported on rollback might be a solution, but hoped there is a more graceful/automated method.