[Bug]: Backups in UI after upgrading to Postgres18.3 give an error

[jsaltzman]

What happened?

Description:
After following the official Postgres 15 → 18.3 upgrade guide, the built-in backup feature fails with pg_dump process exited with code 1.
The root cause is that the sparkyfitness_server container ships with pg_dump version 15 (from Debian 12), but the database is now running Postgres 18.3. Postgres requires pg_dump and the server to be on the same major version.

Environment:

SparkyFitness server image: codewithcj/sparkyfitness_server:latest
Postgres image: postgres:18.3-alpine
Deployed via Docker Compose

Steps to reproduce

Follow the official Postgres 15 → 18.3 upgrade guide
Attempt to run a backup from the SparkyFitness UI or via the scheduled backup feature

Expected behavior

The service should backup

How are you running SparkyFitness?

Docker Compose

SparkyFitness version

SparkyFitness v0.16.5.6

Which area is affected?

Admin/Backups

Relevant logs or screenshots

[ERROR] Backup failed: Error: pg_dump process exited with code 1
Confirmed root cause:
pg_dump: error: aborting because of server version mismatch
pg_dump: detail: server version: 18.3; pg_dump version: 15.16 (Debian 15.16-0+deb12u1)
Expected behavior: The server image should ship with a pg_dump version that matches the supported Postgres version (18.3).
Workaround: Run manual backups from the host using:
bashdocker exec -i sparkyfitness-sparkyfitness-db-1 pg_dumpall -U sparky > ~/sparkyfitness_backup_$(date +%Y%m%d).sql

Anything else?

No response