Develop schema changes
You changed models in the Prisma schema and need corresponding SQL migration files applied locally.
Prisma · Command
prisma migrate dev
prisma migrate dev creates and applies migrations while developing. It uses a shadow database to detect drift and requires a development environment.
How it works
Run this development-only command after changing the Prisma schema. Prisma checks migration history in a shadow database, detects drift, creates a migration for schema changes, and applies unapplied migrations to the development database while updating migration state.
Syntax
prisma migrate dev [options]Parameters
| Name | Type | Description |
|---|---|---|
--url | option | Override the datasource URL from the Prisma config file. |
--config | option | Custom path to a Prisma config file. |
--schema | option | Custom path to the Prisma schema. |
--name, -n | option | Name the migration; -n is the short form. |
--create-only | option | Create a migration without applying it. |
--help, -h | option | Display help information; -h is the short form. |
Returns
Creates migration files and applies applicable migrations; the command reports migration status and updates the _prisma_migrations table.
Key points
- It both generates migrations from Prisma schema changes and applies them during development.
- Prisma replays existing migrations in a shadow database to identify drift.
- Pending migrations are applied to the shadow database before new changes are evaluated.
- Schema edits can produce a new migration during the command run.
- Unapplied migrations reach the development database, and Prisma records their status.
- The datasource connection is read from Prisma configuration, such as
prisma.config.ts.
Common mistakes
- Do not use this command in production.
- A shadow database is required for checking migration history and detecting schema drift.
- MongoDB is not a supported datasource for this command.
Example: Create a migration for review
npx prisma migrate dev --name add-profile --create-onlyCreates a named migration from schema changes without applying it; review the generated SQL before applying it in development.
When to use it
Prepare a migration without applying
You need to inspect or edit generated SQL before applying it.
Check migration consistency
You want to validate migration history and detect drift against a development database.
Interview questions
QWhat does prisma migrate dev do?
It replays migration history in the shadow database to detect drift, applies pending migrations there, generates a migration for Prisma schema changes, and applies unapplied migrations to the development database. It also updates the _prisma_migrations table.
Follow-up: Why use a shadow database? It provides a separate environment where Prisma can replay migration history and compare the resulting schema with the expected state without modifying the development database.
QHow can you create a migration without applying it?
Use --create-only to generate a migration without applying it. This is useful when the SQL needs review or manual editing before application.
Follow-up: How would you supply a migration name? Add --name, or its short form -n, followed by the name, such as --name add-profile.
QWhy should this command not be used in production?
It is intended for development environments and requires a shadow database. Production deployments should use a production migration workflow rather than this command.
Follow-up: What can cause a reset prompt? Schema drift or a migration-history conflict can cause Prisma to request a database reset to synchronize state.
Related
prisma migrateprisma migrate deployprisma migrate resetprisma db pushshadow databasePrisma schema