How the bug manifests and where it came from
Microsoft confirmed that Windows updates released since the October 2026 Patch Tuesday are causing SQL Server performance and stability problems. According to a Windows release health dashboard update, the fault arises when server queries leave connections open after the queries complete. Instead of closing, those connections accumulate and degrade server performance over time.
In the company’s words, "In affected environments, administrators might observe an increasing number of open connections to external databases, even when the associated SQL Server linked server queries are no longer running." The problem is specifically tied to SQL Server linked server queries that use OLE DB providers to connect to external databases.
Scope: affected systems, providers, and behaviors
Microsoft described the issue as impacting both client and server Windows releases. Client systems named as affected are Windows 10 and Windows 11; server systems are Windows Server 2012 and later. The vendor explicitly called out Oracle and PostgreSQL connections among those known to be affected, indicating the issue crosses multiple external database technologies accessed through OLE DB providers.
The observable symptom is a growing number of open external connections even when corresponding linked server queries have completed. Over time, those open connections can impair performance and raise stability concerns on SQL Servers running on impacted Windows builds.
Audit-ready is a season. It shouldn't be.
Evidence in spreadsheets, controls drifting between audits, frameworks multiplying on flat headcount. Nubivance runs continuous compliance on Rapid7 Cyber GRC - SOC 2, HIPAA, ISO 27001, PCI, CMMC.
End the scrambleShort-term mitigations Microsoft recommends
Microsoft told administrators they can manually close connections left open by the bug or restart the SQL service to temporarily address performance and stability problems. The company cautioned those steps are imperfect remedies: restarting the SQL service can kill active queries, and the underlying behavior will resurface after some time.
To provide an interim path, Microsoft said it will deliver an out-of-band Windows update "over the next few days" to fix the issue. While the company works on that update, IT administrators are advised they can contact Microsoft Support for Business for a mitigation that can address the problem until the formal fix ships.
Related Windows update problems and recent history
Microsoft also confirmed two additional, related update issues in recent weeks. First, the company in September published a temporary workaround for SQL Server crashes on some Windows Server 2025 systems tied to recent memory management changes. That bug affects SQL Server when the Lock Pages in Memory (LPIM) Windows policy is enabled and "may cause SQL Server crash dumps, SQL Server services to stop or restart unexpectedly, failures during database maintenance operations, memory corruption errors, and access violation exceptions (with 0xC0000005 error codes)." The earlier memory-management regression and the current linked-server connection regression are separate faults but both arise in the context of recent Windows changes interacting with SQL Server configuration.
Second, Microsoft said yesterday that November 2026 security updates are failing to deploy on some Windows 11 systems where the October preview update has been installed. Affected users report update failures with error codes 0x80070306, 0x800f0989, or 0x800f0993.
What this means for database administrators, IT administrators, and enterprise operations
- Database administrators: Expect to see unexplained increases in external connections on affected SQL Servers, particularly where linked server queries use OLE DB providers to Oracle or PostgreSQL. Manual connection cleanup and service restarts can reduce immediate pressure but may interrupt in-flight work and are temporary.
- IT administrators and support teams: You have a short window to apply Microsoft’s forthcoming out-of-band update or to obtain a vendor mitigation via Microsoft Support for Business. Track deployment status closely on Windows 10/11 and Windows Server 2012+ hosts and be prepared to manage user impact if service restarts are required.
- Enterprise operations and change managers: Coordinate any restarts or manual mitigation steps with application owners. Because the bug can return after temporary measures and because service restarts can kill active queries, schedule intervention to minimize production disruption.
Microsoft’s timeline is concise: an out-of-band fix will be issued within days, and support channels can provide interim mitigations. Until that patch is applied, teams should monitor open external connections on SQL Server instances, weigh the operational costs of manual restarts, and escalate to Microsoft Support for Business if needed.
