My Postgres is crash-looping with "No space left on device" on pg_wal. Disk filled up during an index build. Resized volume 10GB→20GB but filesystem never picked it up. Restored from backup which created a new volume, but Postgres still mounts the old full volume on every restart. WAL replay completes successfully but crashes writing the checkpoint. Data is intact — purely a disk space issue. Old Volume is always mounted despite the restore. Tried: live resize, redeploy, multiple restores — nothing worked. Note: Railway deployment outage (incident QVJT4QJO) occurred at the same time. Could this have caused the volume resize to not propagate? FATAL: could not write to file "pg_wal/xlogtemp.30": No space left on device. @Railway 

Note: there are currently 3 volumes visible — postgres-volume (original), and two dated copies from previous restore attempts. The original postgres-volume is the one mounted to the service and contains all the data. Please expand the filesystem on that one only