BugZero found this defect 1308 days ago.
Data sources
All data on this page is proprietary to BugZero® or gathered from public sources
7/21/2023
SQL Server 2017 on Windows (all editions)
build lower than 14.0.3006.16
14.0.3006.16
Consider the following scenario: You have an AlwaysOn Availability Group (AG) in SQL Server 2017 with database Distributed Transaction Coordinator (DTC) support enabled. During Availability Group failovers, SQL Server internal DTC State for all the databases in the AG is first stopped, and then all the active transactions for all the databases of this availability group are killed. However, before the active transactions are killed, if these transactions try to access the stopped DTC State, a memory corruption may occur. This happens because the transactions are trying to access the memory that has already been deleted.
This issue is fixed in the following cumulative update for SQL Server: Cumulative Update 1 for SQL Server 2017