FIX: Unnecessary failovers occur when a SQL Server 2017 Failover Cluster Instance or Always On AG is managed by Pacemaker

Applies to: SQL Server 2017 DeveloperSQL Server 2017 EnterpriseSQL Server 2017 Enterprise Core More

Symptoms


Assume that you use a Failover Cluster Instance (FCI) or an Always On High Availability Group (AG) in Microsoft SQL Server 2017. If the FCI or AG is managed by Pacemaker, the primary replica may fail over when sp_server_diagnostics reports that some components are in a warning state. For example, this issue may occur during long-running queries or if there are any recent nonfatal process dumps. Only components in error state should cause a failover of the FCI instance or the AG primary replica.

Resolution


This fix is included in the following update for SQL Server:

Cumulative Update 8 for SQL Server 2017

Status


Microsoft has confirmed that this is a problem in the Microsoft products that are listed in the "Applies to" section.

References


Learn about the standard terminology that Microsoft uses to describe software updates.