Local backup disks fill up fast. On a server with 400 accounts, a week of full daily backups can pass 2 TB before anyone notices. Object storage such as Backblaze B2, Wasabi or any S3-compatible bucket costs a fraction of NVMe and survives the loss of the server itself.
What to change first
- Switch from full to incremental backups in WHM » Backup Configuration.
- Keep 7 daily, 4 weekly and 3 monthly copies instead of 30 daily ones.
- Send copies off the server over an encrypted channel, using a key scoped to one bucket.
- Exclude cache folders and node_modules from account backups.
Check disk use before you change anything
$ du -sh /backup/* | sort -h | tail -5
$ df -h /backupTest the restore, not just the backup
A backup you have never restored is a guess. Restore one account to a staging server every month and time it. If it takes longer than your SLA allows, shorten the retention chain or split very large accounts.
Per-account backups to S3, Backblaze B2 and Wasabi