Microsoft Releases Cumulative Update 7 for SQL Server 2019, Including Important Fixes.. BUT…
Microsoft has released Cumulative Update 7 for SQL Server 2019, which includes several important fixes for users. However, shortly after its release, Microsoft removed the update due to a problem affecting snapshots and CHECKDB.
One of the notable fixes included in the update is the addition of a new message in sys.messages. Message 47149 warns users that they cannot start the job on the secondary replica of a contained availability group, and that they should start the job on the primary replica instead. Contained availability groups refer to availability groups in Big Data Clusters that can contain system databases, rather than traditional contained databases with Always On availability groups.
Other fixes included in Cumulative Update 7 address issues such as concurrent inserts against tables with columnstore indexes causing queries to hang, SELECT queries returning an incorrect number of rows when run in Read Committed Snapshot Isolation on a Clustered Columnstore Index, and inconsistency occurring when ghost rows are inserted into mapping index rowset.
Additionally, the update includes fixes for intermittent Availability Group failover, long waits with DTC_STATE wait type in distributed transactions, and several other issues that could impact users.
While Microsoft has temporarily removed Cumulative Update 7 due to an issue affecting snapshots and CHECKDB, SQL Server users should still plan to install the update once it becomes available again. It’s important to carefully review the release notes (https://support.microsoft.com/en-us/topic/kb5004242-cumulative-update-7-for-sql-server-2019-94600026-955a-4e37-b3cb-680bbd405b26) and any available KB articles to understand the changes included in the update and determine the best course of action for your specific environment.
Overall, Cumulative Update 7 for SQL Server 2019 includes several important fixes that address issues reported by users. With work continuing on contained availability groups in Big Data Clusters, users can look forward to additional support for these types of availability groups in the future.