Database migrations
As described in the upgrade guide, you should use the
or
command to update the database.
We use database migrations in order to handle changes to the database schema in a more secure and stable way. This is actually very common. The thing is that when just changing the schema in schema.prisma
without creating migrations, the update to the newer database schema can damage or delete all data in production mode, since the system sometimes doesn’t know how to transform the data from A to B. Using migrations, each step is reproducable, transparent and can be undone in a simple way.
Creating migrations
If you are modifying the codebase and make a change to the schema.prisma
file, you must create a migration.
To create a migration for your previously changed schema.prisma
, simply run the following:
Now, you must create a short name for your migration to describe what changed (for example, “user_add_email_verified”). Then just add and commit it with the corresponding code that uses your new database schema.
Always keep an eye on what migrations Prisma is generating.** Prisma often happily will drop entire columns of data because it can’t figure out what to do.
Error: The database schema is not empty
Prisma uses a database called _prisma_migrations
to keep track of which migrations have been applied and which haven’t. If your local migrations database doesn’t match up with what’s in the actual database, then Prisma will throw the following error:
In order to fix this, we need to tell Prisma which migrations have already been applied.
This can be done by running the following command, replacing migration_name
with each migration that you have already applied:
You will need to run the command for each migration that you want to mark as applied.
Resetting Prisma migrate
When your local Prisma database runs out of sync with migrations on local and you are tearing your hair out, I’ve been there, so you don’t have to:
PostgreSQL
Quickly re-index
Was this page helpful?