SQL Server 2019/2022 crash EXCEPTION_STACK_OVERFLOW during master DB startup on new laptop hardware

Question

Hardware: HP Laptop 15-fd1xxx (new, purchased recently)

CPU: 1 socket, 9 cores, 18 logical processors

RAM: 24 GB

OS: Windows 10 Home, Build 26200 (stable channel, not Insider)

Issue:

SQL Server fails to start on every attempt, both with SQL Server 2019 (RTM 15.0.2000.5)

and SQL Server 2022 (RTM 16.0.1000.6). The crash is 100% reproducible and occurs

at the exact same point every time: right after buffer pool allocation and

soft-NUMA activation, during master database startup.

SQL Server 2019 first attempt threw:

• Exception code: 0xc0000005 (access violation) in ntdll.dll

Subsequent attempts (both 2019 and 2022) consistently throw:

• EXCEPTION_STACK_OVERFLOW (0xc00000fd)
• Stack Signature: 0xBDDB9855 (identical across every crash)

Troubleshooting already performed, no effect on the crash:

• Disabled Memory Integrity / HVCI (Core Isolation)
• Disabled third-party antivirus (McAfee) real-time protection
• Windows memory diagnostic: passed, no errors
• Windows fully updated (including July 2026 cumulative update)
• BIOS updated to latest version via HP Support Assistant
• Not on Windows Insider channel
• Disabled Virtualization Technology in BIOS (confirmed via ERRORLOG, "(Hypervisor)" tag no longer present) — crash still identical
• No accessible P-core/E-core control in this BIOS to test further

ERRORLOG excerpt (SQL Server 2022):

[Server] Automatic soft-NUMA was enabled because SQL Server has detected

hardware NUMA nodes with greater than 8 physical cores.

[spid] Starting up database 'master'.

[spid] Stack Overflow Dump not possible - Exception c00000fd

EXCEPTION_STACK_OVERFLOW

[spid] Stack Signature for the dump is 0x00000000BDDB9855

Question: has anyone seen this specific stack signature or a similar

crash pattern on hybrid-core CPUs? Is there a known compatibility issue

between SQL Server RTM builds and this CPU/NUMA configuration, and if so,

is there a fix or workaround (e.g. specific CU, trace flag, or startup

parameter) short of running SQL Server inside a VM?

(Question written thanks to Claude, I'm not that an expert)

Answer 1

The AI was right. There is this message:

There have been 256 misaligned log IOs which required falling back to synchronous IO. The current IO is on file C:\Program Files\Microsoft SQL Server\MSSQL15.ELLI_DB\MSSQL\DATA\master.mdf.

This is due to a known issue where Windows 11 reports a disk-sector size that SQL Server does not support. Microsoft has published the article Troubleshoot SQL Server errors related to system disk sector size greater than 4 KB which discusses workarounds for this problem. Basically, you need to make the registry edit described in the article. Very important! You need to reboot Windows for the change to take effect.

Answer 2

This pattern matches a known Windows 11 disk sector-size compatibility issue, not a documented hybrid-core or soft-NUMA compatibility problem in the provided material.

The strongest indicator is the failure during master startup together with the setup/startup symptom "Wait on the Database Engine recovery handle failed" and the SQL Server error log pattern:

There have been 256 misaligned log IOs which required falling back to synchronous IO.

That issue is documented for systems upgraded to or running on Windows 11 where some drives report a sector size that SQL Server cannot handle correctly. It can prevent SQL Server from starting and can surface during master.mdf recovery/startup.

What to do:

1. Open the SQL Server error log and check for the misaligned I/O message.
   • Typical path for a default SQL Server 2022 instance: C:\Program Files\Microsoft SQL Server\MSSQL16.MSSQLSERVER\MSSQL\LOG\ERRORLOG
   • For a named instance, the MSSQL16.MSSQLSERVER portion differs.
2. If the error log contains the 256 misaligned log IOs message, treat this as the known disk sector-size issue.
3. Apply the documented workaround for the sector-size problem.
   • In the related Q&A guidance, the recommended startup parameter is -T1800.
   • Add it in SQL Server Configuration Manager > SQL Server Services > <instance> > Properties > Startup Parameters, then restart the service.
4. If this is happening during setup and setup reports Wait on the Database Engine recovery handle failed, the same underlying startup problem can be the cause. The next step is still to inspect ERRORLOG for the misaligned I/O entry.
5. If the misaligned I/O message is not present, the provided material does not identify a supported fix for the specific stack signature 0xBDDB9855, hybrid-core CPUs, or a soft-NUMA-specific defect.

Based on the available material, there is no documented fix here that ties this crash to hybrid-core CPUs, automatic soft-NUMA, or a specific SQL Server RTM access violation/stack overflow on that CPU topology. The documented and repeated startup failure pattern instead points to the Windows 11 sector-size issue during master database startup.

---

• Troubleshoot SQL Server errors related to system disk sector size greater than 4 KB
• SQL Service not started even SQL installed successfully - Microsoft Q&A Issue Summary for HP Support – SQL Server Stack Overflow on HP EliteBook
• After upgrading to Windows 11, the Sql Server crashes. - Microsoft Q&A I had originally a Sql Server 2019 version. After upgrade, it didn't start. I uninstalled it and installed Sql Server version 2022. But this version is also unable to start.
• TITLE: Microsoft SQL Server 2022 Setup ------------------------------ The following error has occurred: Wait on the Database Engine recovery handle failed. Check the SQL Server error log for potential causes. For help, click: https://go.microsoft.com/f - Microsoft Q&A
• Need expert help with sql server problem - Microsoft Q&A Unable to install default instance of SQL server 2019 Getting error stating Database recovery handle has failed SQL ERRORLOG