Consider the following scenario:
- You have a readable secondary replica of Microsoft SQL Server 2016 AlwaysOn Availability Groups (AG).
- You use parallel redo instead of single thread redo.
In this scenario, you will notice that assertion occurs with a memory dump, and data movement may be suspended. Additionally, you may receive an error message that resembles the following:
Microsoft has confirmed that this is a problem in the Microsoft products that are listed in the "Applies to" section.