We have a read-only replica of our production database using Availability Groups (SQL Server 2017). The requirement was to set the recovery model to full for both.
Is there a reason why this is a prerequisite? I get that the primary database should have a full recovery mode for the changes to be replicated, but why does the read-only replica of it require it?
I did some research and I saw that Database Mirroring requires Full Recovery Model as well.
The problem that we have is that we have some queries in millions of rows that fill up the tempDB and our DBAs complain. I thought that switching the read-only replica to a simple recovery model (since we are not going to need or use restore point in time on the replica instance) would resolve the issue but apparently this is impossible.
Could someone explain the requirement or suggest a workaround?