I need help with a serious incident in my database.
Accelerate
Prisma Postgres💻CLII had a database with approximately 800,000 records. This morning, a drift issue appeared during migrations using Prisma, as the Antigravity AI had executed a database push. When I tried to revert it because it had been applied incorrectly, the schema ended up breaking.
The system suggested performing a database reset, which was clearly not feasible given the amount of stored data.
I tried to find an alternative and started working with migration baselines. During that process, it prompted me to configure a shadow database URL, so I temporarily changed the Prisma connection URL.
After that, the problems began: some users notified me that tables no longer existed. Upon directly checking the database, I confirmed that all the tables were gone.
I checked the backups and found one generated around 10:00 AM. I tried to restore it in a recovery environment, but the restore failed and did not complete, displaying an error during the process.
Has anyone experienced something similar with Prisma, migration drift, or backup restoration? Any guidance or experience you can share would be greatly appreciated.
The system suggested performing a database reset, which was clearly not feasible given the amount of stored data.
I tried to find an alternative and started working with migration baselines. During that process, it prompted me to configure a shadow database URL, so I temporarily changed the Prisma connection URL.
After that, the problems began: some users notified me that tables no longer existed. Upon directly checking the database, I confirmed that all the tables were gone.
I checked the backups and found one generated around 10:00 AM. I tried to restore it in a recovery environment, but the restore failed and did not complete, displaying an error during the process.
Has anyone experienced something similar with Prisma, migration drift, or backup restoration? Any guidance or experience you can share would be greatly appreciated.
