DOCUMENTATION · VERSION 1.0
SaintBackup collects MariaDB backups from many servers onto one management machine, encrypts them end to end, and runs a restore test every day to prove each backup works.
20 articles, in the order you'll actually do things: install, connect, back up, restore, operate.
Can't find what you need?
Paste the full log line into the search box above — articles in the Troubleshooting guide are indexed by the exact error message.
Back to the search box ↑From download to your first backup that has passed a restore test. Each step links to a detailed article.
On your management machine: an office server, a VPS, or a laptop. Activate the 1-month free trial.
Install the Center →On the machine running MariaDB (Windows or Linux). The Agent holds the database credentials, and you choose which databases to share.
Install the Agent →Pick one of the connection methods, enter the 8-character pairing code, then compare the 6-digit confirmation number on both machines.
Pairing →Choose databases/tables and a schedule (daily full, incremental every 30 minutes by default), then check your first restore test.
Choosing databases/tables and schedules →13 real errors encountered while running mariabackup and SaintBackup. Each article starts with the exact log line, followed by the cause and the fix.
[ERROR] InnoDB: Unsupported redo log format.
The redo log was created with MariaDB 11.4.Cause: the mariabackup used to back up or prepare is a different version from the MariaDB server that wrote the redo log.
Fix: let the Agent and the Center pick the matching mariabackup automatically; if you prepare manually, use the build that matches the source server's version.
Read the full article →ERROR 1213 (40001): Deadlock found when trying
to get lock; BACKUP STAGE BLOCK_COMMITCause: a known MariaDB 10.5.8 bug when switching BACKUP STAGE phases while transactions are running.
Fix: SaintBackup retries automatically, so usually no action is needed. If it keeps happening, upgrade MariaDB.
Read the full article →ERROR 1227 (42000): Access denied; you need
(at least one of) the RELOAD privilege(s)Cause: the database account held by the Agent hasn't been granted the RELOAD privilege.
Fix: grant the backup account the privileges listed in the Install the Agent article, then run it again.
Read the full article →request rejected: clock skew 412s > 300sCause: every request is signed with a timestamp; if the clocks differ by more than 300 seconds, the receiving side rejects it.
Fix: turn on automatic time sync (NTP) on both the Agent machine and the Center machine.
Read the full article →incremental chain broken: from_lsn does not
match to_lsn of previous backupCause: an incremental backup in the chain is missing or corrupted, so the chain can't be continued.
Fix: run a new full backup; subsequent incrementals will chain from that full backup.
Read the full article →8 more errors in the guide
Connection drops mid-transfer, the Center being offline, MyISAM tables, failed restore tests, and other common errors.
Open the Troubleshooting guide →When a backup or restore test fails, the assistant will read the log from that specific run, check it against the Troubleshooting guide and the docs, and suggest concrete steps to fix it.
Not available in version 1.0. In the meantime, the search box above accepts full log lines.