My backup finished with warnings or shows Partial. What does it mean?
A Partial point or a Warnings badge means BackupWall backed up most of the data but could not read or store something. Open the backup point and read the message: it names the database, file or section. Fix the cause and the next run tries again.
Where to look
- Open Backup points and choose the run.
- Read the message and the Warnings badge. The jobs list shows "Partial: N items need attention"; hover for the reason.
- Check
/var/log/backupwall/worker.logfor detail.
Common messages and what to do
| Message | Cause | What to do |
|---|---|---|
| Databases not backed up: ... (Partial) | A database dump failed after its retry, or its upload failed. A dump that ended early (cut stream, missing "Dump completed" line) is retried. | Read which database and why. Check that database is not corrupt or huge, that the database server is running, and the disk space in /var/lib/backupwall. A dropped database is not removed from later points. |
| Databases (the panel could not list them ...) | The panel or database server did not answer when listing the account's databases. | Fix the panel or database connection and run the job again. |
| Files not backed up (they exist but could not be read or stored; ...) | A file or folder could not be read (permissions, a vanished mount, a bad path). | The point keeps the older stored version of that file, if any. Fix the permissions and the next run tries again. Up to five examples are listed per account. |
| ... outside the account home | A file was reached through a link or a swapped directory that leaves the account's home. BackupWall refuses to read outside the home. | Back up that target with a Directory job. |
| PostgreSQL database: ... | A PostgreSQL database could not be dumped, or PostgreSQL could not be reached. | Check that PostgreSQL is running and reachable; databases on a remote server are skipped on Plesk. |
| Not in this backup: ... (on a restore) | The point was taken by an older build before a section was captured. | Nothing is wrong. The restore leaves the account's current settings for that section alone. Run a fresh backup. |
| Interrupted: the worker process restarted or crashed before this run finished | An upgrade, a restart or a crash stopped the run. | Nothing starts it again by itself: the job's next scheduled run or Run now takes over and reuses what was already uploaded. |
| A section failed to capture (DNS zones, SSL certificates, FTP logins, cron jobs, mail settings) | The panel refused a read. | The point names the gap, so a failed read never looks like "none". Fix the panel and run again. |
Get warnings sent to you
Warnings arrive as a "partial" alert. Email shows "completed with errors"; Slack, Discord and webhook channels get it through the Backup failed event. See Alerts and notifications.
Other things that look like warnings
- A file that changes while it is read (a mailbox, a SQLite database) can be stored torn; the next run re-uploads the parts that changed. Use logical dumps, or stop the writer.
- On MyISAM and Aria tables, dumps with
--single-transactioncan be inconsistent. - Symlinked data is not followed. Add a Directory job for the target.
Still stuck? Open a ticket at portwall.net/billing or write to [email protected], and attach a support bundle. See Report a problem and make a support bundle.