Question

Does anyone know what would be the cause of inflating database backup sizes after adding a database to a high availability group? I've noticed this behavior with not only the example below but any database I've added to the HA group. I don't think I can support the HA option in my environment if this is the intended behavior.

Database File Size 322560.00 MB; Space Available 23444.92 MB

Full backup size with compression before HA group: 4188.3 MB

Full backup size with compression after adding to HA group: 48697.4 MB

Full backup size with compression after removing from HA group: 4925.1 MB

Full backup size with compression after re-adding to HA group: 48732.2 MB

HA group is in a healthy state with these settings: Asynchronous, Manual Failover, Allow all Connections in Primary Role, Readable in Secondary, Standard port 5022 for endpoint, Primary only for backup preferences. Only syncing to one server at a DR state in another state.

Backup routine consists of FULL every weekend, DIFF every week night, LOG every hour and databases not in the HA group still have our standard compression ratio with that same MA plan.

Command in our custom log is: BACKUP DATABASE [DATABASE] TO DISK = N'H:\DB\MSSQL11.MSSQLSERVER\MSSQL\Backup_Remote\DATABASE.bak' WITH CHECKSUM, COMPRESSION

Was it helpful?

Solution

The log is part of the backup. I suspect that you are pulling in a large log when performing the backup while the database is part of an AG. You can confirm this by comparing the file sizes when each backup takes place, or after the fact by performing a RESTORE FILELISTONLY for each backup. I suspect you'll see very different sizes for the log files.

Licensed under: CC-BY-SA with attribution
Not affiliated with StackOverflow
scroll top