I keep getting this error, even if I kill all pg backends and re-run the migration

'Context: Timed out trying to acquire a postgres advisory lock (SELECT pg_advisory_lock(72707369)). Elapsed: 10000ms. See <https://pris.ly/d/migrate-advisory-locking> for details.',

It could be a couple of things: • There is a migrate process still running somewhere that is holding the lock. • If you see this error while deploying the app, it probably means there are multiple deployments running concurrently. In that case, the error is expected for all but one of the runners.

I terminated the pg backends and waited a little while and was able to run the migrations again later on. It was because a migration failed. I had to go manually update the

_prisma_migrations

table to “roll back” the migration. Would be great to find a nicer way to do this in serverless-aurora-land