Backup failed: how to read backup and restore errors

For server administrators · Updated 2026-10-29

On this page
  1. Symptoms
  2. Where to find the full error
  3. Causes and fixes (most likely first)
  4. Still not working?

Symptoms

  • On Backup Jobs, the Last run column shows a red failed count, with up to five accounts and their error below it.
  • In Daily Snapshots or Full Backups, a copy has the status failed with a short error under it.
  • A full backup shows external copy failed (hover over it to see why).
  • A restore did not change anything, or the backup shows an error starting with "Restore failed:".

Where to find the full error

  1. Backup Jobs (section Websites) → Last run: the first 120 characters of each error.
  2. Daily Snapshots / Full Backups → Status column: the error of that copy.
  3. Task Queue in the menu: every backup and restore is a task (snapshot.create, backup.create, snapshot.restore, backup.restore). Open a failed task for the whole message, then click Retry after fixing the cause.

Causes and fixes (most likely first)

"External storage: …"

The external destination refused or could not be reached. The rest of the message comes from the storage. Fix: open Backup Jobs, click Test next to the destination, correct the host, keys, password, bucket or folder with Edit, and save. For Google Drive, OneDrive or Dropbox, click Connect another account and sign in again. The archive of a failed upload is kept on this server and can still be restored from Full Backups.

The disk is full (the error mentions "No space left on device")

Backups need free space in /backup, also for jobs to external destinations (each archive is built there first). Fix: free space as described in how backups use disk space, then run the job again.

"Account no longer exists"

The account was deleted between queueing and running the backup. Nothing to fix.

"A backup or restore is already in progress for this account." / "A snapshot of this account is already running."

Another backup or restore of the same account is running. Wait for it to finish. Scheduled jobs skip such an account automatically.

"This copy has no files (its job copies databases or email only)." / "This backup has no files …"

You asked to restore files from a copy that holds only databases or email. Choose Don't restore files, or pick a copy from a job that copies Files.

"Only completed backups can be restored." / "The account no longer exists. Re-create it first, then restore."

The backup is not done, or its account was deleted. For a deleted account see backups of deleted accounts.

"The backup destination no longer exists"

The archive was at a destination that has since been removed, so it cannot be fetched. Add the destination again with the same settings, or download the archive from the storage yourself.

"Database …: …"

The database server refused the import during a restore. The text after the database name says why (often a full disk).

You selected a symbolic link. Restore the real folder or file instead.

Jobs do not run at all

Tasks waiting longer than 2 minutes show a warning at the top of the admin panel: the background service is not running. Check it on the server:

systemctl status sahaabd

Also check that the job is not marked Paused on Backup Jobs. If it is, open the job, tick Active (runs on its schedule) and click Save.

Still not working?

Open the failed task in Task Queue and use Fix with the assistant or Explain with AI if they are shown, or send the full error text to your support contact.

Was this article helpful?

Still stuck?

Ask the AI about your exact situation — it answers from this knowledge base — or contact us.