Alpha database migration guide

D1’s open beta launched in October 2023, and newly created databases use a different underlying architecture that is significantly more reliable and performant, with increased database sizes, improved query throughput, and reduced latency.

This guide will instruct you to recreate alpha D1 databases on our production-ready system.

​​ 1. Verify that the database is alpha

$ npx wrangler d1 info <database_name>

​​ 2. Create a manual backup

$ npx wrangler d1 backup create <database_name>

​​ 3. Download the manual backup

The command below will download the manual backup of the alpha database as .sqlite3 file:

$ npx wrangler d1 backup download <database_name> <backup_id> # See available backups with wrangler d1 backup list <database_name>

​​ 4. Convert the manual backup into SQL statements

The command below will convert the manual backup of the alpha database from the downloaded .sqlite3 file into SQL statements which can then be imported into the new database:

$ sqlite3 db_dump.sqlite3 .dump > db.sql

Once you have run the above command, you will need to edit the output SQL file to be compatible with D1:

1. Remove BEGIN TRANSACTION and COMMIT; from the file.

2. Remove the following table creation statement:

   CREATE TABLE _cf_KV (
    	key TEXT PRIMARY KEY,
    	value BLOB
   ) WITHOUT ROWID;
   

​​ 5. Create a new D1 database

All new D1 databases use the updated architecture by default.

Run the following command to create a new database:

$ npx wrangler d1 create <new_database_name>

​​ 6. Run SQL statements against the new D1 database

$ npx wrangler d1 execute <new_database_name> --remote --file=./db.sql

​​ 7. Delete your database

To delete your database, run:

$ npx wrangler d1 delete <database_name>