Skip to content
Docs

Scheduled Purge

Automatically delete files older than an age you set, for retention cleanup of stale exports, temporary files, and logs. It permanently deletes files — and is not a backup.

Last updated July 17, 2026

#What Scheduled Purge does

A Scheduled Purge is a job mode — chosen from the same list as Copy, Move, and Sync/Mirror — that deletes files older than a configured age from a single location, optionally limited to a file pattern, on a schedule. It's built for retention cleanup: stale exports, temporary or already-processed files, vendor drops, and old generated reports or logs. Scheduled Purge is available on Business plans and above (Business, Business Premium, and Enterprise).

Scheduled Purge permanently deletes matching files and is not a backup. Deleted files are gone unless you have a separate backup or archive. Treat it as housekeeping, never as a retention store.

#How it decides what to delete

A purge job has no destination. You configure:

  • Target — one location (Local, SMB, SFTP, FTP, S3, or Azure Blob) and a folder or prefix inside it. Remote targets use a stored credential with list and delete rights.
  • Age — “delete files older than” a number of days, weeks, months, or years. Files newer than the cutoff are never touched. (Months and years use calendar-aware math, so “1 month” isn't always 30 days.)
  • Pattern — a file pattern such as *.csv targets specific files; the default * matches everything by age.

Both the age and the pattern must match for a file to be deleted.

#Built-in safety

FileAgent includes guardrails to reduce the chance of a damaging purge:

  • It refuses to purge the bare root of a drive, share, bucket, or container — point the job at a folder inside the location.
  • It never deletes files newer than your cutoff.
  • If a file's age can't be determined, it is skipped rather than deleted (fail-safe).
  • You can archive before delete, keeping a recoverable copy of what was purged (see Archive options).

#Preflight checklist before production

  1. Point the job at a specific folder/prefix, not a root.
  2. Start with a narrow pattern (for example *.tmp), not the default *.
  3. Confirm the age threshold against real files' modified times.
  4. Decide whether to archive before delete, or ensure a separate backup exists.
  5. Run it manually once and review the deleted and skipped counts in Logs.
  6. Only then enable the schedule.
Run a purge manually and review the run detail before scheduling it. The run records what was scanned, deleted, and skipped — your fastest way to confirm the rule does what you intended.

#If it deletes too much or too little

See Scheduled Purge troubleshooting — the most common causes are a pattern that's broader or narrower than intended, an age threshold that's larger or smaller than you meant, or files that simply aren't old enough yet.