5 days ago
Project: generous-prosperity / production. Service: Postgres-7FXU (+ wa-ai-crm). Volume: postgres-volume (postgres-volume-O7HU).
The volume was 500 MB and filled up. I resized it to 5 GB with Live Resize (Settings shows 5 GB), but Metrics still shows ~470 MB capacity at 98%, and Postgres crashes during WAL recovery with “could not write to file pg_wal/xlogtemp: No space left on device”. This looks like the same issue you fixed in another thread: the volume is 5 GB at the ZFS level but the ext4 filesystem was never expanded. Please expand the filesystem to the full 5 GB and redeploy Postgres. This is my production database, so please do not wipe the volume. Service IDs: [paste from the service/volume Settings pages].
1 Reply
5 days ago
We checked the volumes on your project: the volume actually attached to Postgres-7FXU at /var/lib/postgresql/data is postgres-volume-O7HU, and it is still provisioned at 500 MB. The 5 GB resize landed on a different volume, postgres-volume, which is not attached to any service, so Postgres is still running out of space on the original 500 MB disk during WAL recovery.
To fix it, resize postgres-volume-O7HU itself to 5 GB (the maximum on your current plan) from its volume settings, then redeploy Postgres-7FXU. Growing a volume does not wipe its data.
Status changed to Awaiting User Response Railway • 5 days ago