Question

Postgres primary. Disk usage on the data volume climbed steadily over 2 days and just hit 100%; the DB is now refusing writes / crashing on `could not write to file ... No space left on device`. The `pg_wal` directory is huge. There was no traffic spike. Two weeks ago a team set up a logical-replication subscriber for a new data pipeline; that subscriber has been down for days. Triage, get writes back safely, and prevent recurrence.