We have SQL Server 2016 Web Edition on production currently. Since this edition supports replication as Subscriber only, I set up a new server with SQL Server 2019 Standard edition. I want to configure SQL Server 2019 as Publisher and SQL Server 2016 as subscriber. To initialize data for Publisher database on SQL server 2019, I created a backup on SQL Server 2016 and restored it on SQL Server 2019. Since our database is very large, I tried to initialize replication from a backup. So I did reverse backup-restore again by creating a backup of Publisher on SQL Server 2019 and restoring it on Subscriber on SQL Server 2016. But this did not work because [SQL Server 2019 backups cannot be restored by any earlier version of SQL Server] (https://learn.microsoft.com/en-us/sql/relational-databases/databases/copy-databases-with-backup-and-restore?view=sql-server-ver15). Could you please tell me what is the best method for initializing the replication in this case? Thank you very much for reading my question!
2
-
How large is "very large"? Initializing via Snapshot is usually the best way when going down a version. Do you have specific criteria that leveraging the (default) Snapshot method won't work? Including the pertinent details for why you think that won't work will be very helpful to answers– AMtwoFeb 6, 2021 at 5:48
-
Hi @AMtwo, The DB size is about 70GB. I think Snapshot will take time in this case and I don't want to make our services unavailable for too long so I'm going to initialize using backup.– T.GiangFeb 6, 2021 at 6:51
Add a comment
|
1 Answer
After doing some research I go with Initialize a Subscription Manually