SQL Server 2019 · Release Health
Known issues in SQL Server 2019
Problems Microsoft has confirmed for this version, from its official release health notes. 4 still open or mitigated.
Active known issues
Open & Mitigated-
Linked server queries that use MSDASQL fail with error 7416
Linked-server queries using the
MSDASQLprovider and a provider string might fail with error 7416, reporting that no login mapping exists. The Database Engine's stricter connection validation can reject certain configurations that worked in earlier builds.How to apply the workaround WorkaroundUse the workaround guidance in the linked Microsoft troubleshooting article.
-
CDC capture job might fail if columns are missing in CDC change tables
On databases using change data capture (CDC), columns with commas in their names might be missing from CDC tracking tables and cause capture-job failures; a comma-containing column in a tracking table might be dropped during a CDC upgrade. Separately, CDC upgrades can fail when tracking-table column names contain other symbols.
How to apply the workaround SQL Trace FlagBefore upgrading the GDR build, enable trace flag T15006 and restart the SQL Server service.
Step-by-step workaround instructionsShow steps Hide steps3 stepsStep-by-step procedure- In SQL Server Configuration Manager, open the instance properties and go to the Startup Parameters tab.
- Add trace flag
T15006, select Add and Apply, then restart the SQL Server service. - Perform the GDR upgrade.
-
Issue two: SQL Server VSS Writer might fail to perform a backup because no database is available to freeze
Backup tools might fail to create application-consistent backups because SQL Server VSS Writer has no database available to freeze. The issue is fixed in CU15.
How to apply the workaround WorkaroundThe page's current solution is to roll back to CU13 and perform the snapshot backup.
Step-by-step workaround instructionsShow steps Hide steps2 stepsStep-by-step procedure- Roll back to SQL Server 2022 CU13.
- Perform the snapshot backup.
-
Incorrect behavior of SESSION_CONTEXT in parallel plans
Queries using SESSION_CONTEXT might return wrong results or create access violation dump files when they run in parallel query plans, especially if a session is reset for reuse.
Workaround status No workaround yetMicrosoft has not published a manual workaround for this issue yet. Defer KB5033688 on affected systems until a fix is released.
Recently resolved
Fixed in latest updates-
Issue two: Patching a read-scale availability group (Windows or Linux) causes the availability group on the patched replica to be removed
Patching a Windows or Linux read-scale availability group with CU31 removes its metadata on the patched replica. SQL Server 2019 CU32 fixes this issue.
Resolution Fixed in KB5054833Install KB5054833 (or a later cumulative update) on affected devices. No manual workaround is required after updating.
-
Issue four: Possibility of error 1204 due to disabled lock escalation
Disabled lock escalation can cause error 1204, stating that the SQL Server Database Engine cannot obtain a LOCK resource. SQL Server 2019 CU28 fixes this issue.
How to apply the workaround WorkaroundUninstall CU26.
-
Issue one: Excessive logging of asynchronous API call warning in the error log
The warning about a long asynchronous API call might fill the SQL Server error log with thousands of entries. SQL Server 2019 CU28 fixes this issue.
Resolution Fixed in KB5039747Install KB5039747 (or a later cumulative update) on affected devices. No manual workaround is required after updating.
-
Issue three: Patching error for secondary replicas in an availability group with databases enabled replication or CDC
Setup can initially fail when patching an availability group database with replication or CDC enabled because the catalog upgrade scripts cannot access the database in single-user mode. SQL Server starts again without single-user mode and completes patching; CU16 fixes this issue.
How to apply the workaround SQL Trace FlagBefore running the patch, enable Trace Flag 12347 and remove it after patching, or remove the affected database from the availability group or remove CDC or replication from that database.
Step-by-step workaround instructionsShow steps Hide steps4 stepsStep-by-step procedure- Before running the patch, choose one of the following options.
- Enable Trace Flag 12347, then remove it after patching.
- Drop the CDC- or replication-enabled database from the availability group.
- Remove CDC or replication from the database in the availability group.
No known issue matches these filters.