Backing Up DB with Filesteam

[HansMeiser]

Hello,

currently we use Veeam Backup & Replication 11.

One of our SQL Testserver uses filestream Feature and some filestreamdata. I expect that filestreamdata were backed up correctly, because i can see in Veeam SQL Explorer a section about "BLOB stores" which shows the used filestream-path.
Currently i did not try a restore. Which is the correct way?
When deleting the database and the filestream Path i expect that a restore would bring back all the data and the path. Is this true?
How would it work if i just want to create a copy of the database by restoring using a new name. What happens to the filestreamdata? Using both DBs the same Filepath or is also the filepath duplicated?

Thank you,
Hans

[PetrM]

Hi Hans,

If you're restoring to the original location, all database files including BLOB stores on production will be overwritten by the restored ones. If you're restoring the database to a different location, for instance, it could be the same server but you specify a new database name: you can specify paths for all database files including BLOB stores. For example, you can create a folder directly from the restore wizard for BLOB stores.

Two different databases cannot use the same file stream data, every database has its own set of data files.

Thanks!

[HansMeiser]

Hello Petr,

thanks for your answer. The described behaviour seems to be very logical to me. I can confirm that in backupbrowser i see a new part "blob storage" for my specific DB, so i think backing up was succesful. Unfortunately i could not do a final restore test, because the admin of the application itself was not contactable these days. I think this will happen to the end of the week.

This leads me to a new question: Currently we have SQL Backups configured to be 1 time a day and transaction logs are backupped every hour (kept for 5 days). Are filestream data backed up also only once a day or are they corresponding to transaction log backups? So i think to have a consistent database in every case filestreamdata should be part of transaction logs. right? Or is this solved in an other way?

Thank you,
Hans