The Postgres container has been in a crashloop since 2026-05-20 04:53 UTC.

Last successful checkpoint: 2026-05-19 20:08:18 UTC.

Error in container logs (repeating every ~20 seconds for ~4 minutes,

then container gave up):

ERROR (catatonit:2): failed to exec pid1: No such file or directory

The volume mounts successfully but the entrypoint binary appears to be

missing from the container image. This started without any deployment

or configuration change from my side — the only user action around

that time was triggering a manual backup the day before.

The app service (FastAPI + Alembic) is now also failing because it

cannot connect to the database (asyncpg TimeoutError on every Pre-Deploy

Step attempt).

Production is fully down. I have NOT touched the volume or the service

configuration since the incident — data should be intact on the volume.

Thank you.