A restore ended partial, or restored files are owned by root. Why?
A partial restore means some items failed and the rest was restored: open the restore entry to read why. Files owned by root, or a cron error on a server with no panel, usually mean the account's unix user did not exist. Create it with the same uid and gid and restore that account again.
Read the reason first
Open the restore on the Restore page. "Completed with errors" (partial) lists what failed. For a Restore all accounts run, the run page groups the reasons and tells you what to do; download the summary as CSV or JSON.
Common reasons
| Reason | What it means and what to do |
|---|---|
| Files owned by root, or cron: ... unknown / "unix user does not exist" (no panel) | The account's unix user was missing when the account was restored. Create it with the recorded uid, gid and shell and no home directory (the restore result prints the exact groupadd and useradd commands), then restore that account again with Overwrite. Or turn on Settings → Server → System users → Create missing system users on restore. |
| Database ... errors (no panel) | MariaDB or MySQL is not running or root cannot use it through the local socket. Check mysql -uroot -e 'SELECT VERSION();'. |
| Interrupted: the worker stopped | Choose Retry failed (restore all) or restore again. |
| Refused: the safety copy could not be taken | The account was on the server and was not changed. Fix the cause shown (destination disabled, a backup of that account running), then restore again. See Safety copy and undo. |
| File names too long | Names of about 250 bytes or more back up but cannot be restored. Every other file was. |
| SSL certificates: this server's panel type cannot install them (no panel) | The certificate files come back with the account's files; install them in your web server. |
| A certificate or DNS zone for a domain the account no longer has | Skipped with a note, not an error, on a whole-account restore. |
| A folder where the backup had a file | Restore adds and overwrites, never deletes. That entry fails. Remove the folder and restore again. |
| A single file fails | An immutable (chattr +i) file, a full disk or a quota. |
| "Not in this backup: ..." | The point predates that section. Nothing is wrong. |
| Mailing list missing from a suspended DirectAdmin account | DirectAdmin refuses it while suspended. Switch the account on and restore again. |
| cPanel account could not be made again from its config archive | The restore says so and ends partial. Create the account, then restore into it. |
Ownership notes
Ownership is restored by user and group name. A name that does not exist on the server gets the numeric id the backup recorded (when no one else uses it), else root. The restore result counts and names the owners it could not find. A failed chown is not reported, so fix ownership by hand if needed.
A restore that looks hung
A restore runs as long as it makes progress. One that reads or writes nothing for 2 hours is stopped and tried once more from the beginning. A large database import that moves no data for 1 hour is given up (a very long index build at the end can look like that).