Backend Migrations

Database migrations apply schema changes reversibly by separating structural updates from data transformations. Use versioned migration files with rollback methods to enable zero-downtime deployments, ensuring backward compatibility and allowing you to revert changes if issues arise.

Prisma migrations generate timestamped files in prisma/migrations that track all schema changes version-controlled. Write descriptive migration names, implement down methods for rollbacks, and manage constraints like foreign keys and NOT NULL to maintain data integrity across deployments.

Alter table migrations add or modify columns with proper constraint handling—use nullable columns for existing rows, then tighten constraints in follow-up migrations. Index large tables strategically and separate schema changes from data population to prevent locks and ensure consistency.

Reversible migrations include down methods that undo schema changes, allowing safe rollback if deployment fails. Version control tracks all migration files, letting you revert to previous states and diagnose issues without data loss.

Splitting schema changes from data transformations enables zero-downtime deployments—deploy schema first, then data migrations in background jobs. This prevents application downtime, reduces lock contention on large tables, and lets you abort data operations independently if needed.

Descriptive naming—like `add_status_column_to_users`—documents intent, aids debugging, and helps teams understand schema evolution. Combined with timestamps, it prevents naming conflicts and makes version control history readable across your entire database changelog.