1. Bad or Corrupt Driver (Most Common Cause)
I've seen this error pop up more times than I can count on Windows 10 and 11 machines. The most common trigger? A driver that's either outdated, corrupt, or just plain badly written. Last month, a small accounting firm called me because their file server kept crashing with 0xC0000118 after every third reboot. Turns out, the NIC driver – a Realtek PCIe GbE – had a known bug in version 10.0.0.12. Rolling it back fixed it instantly.
How to Fix It
- Boot into Safe Mode (hit F8 or Shift + Restart from the login screen).
- Open Device Manager (right-click Start > Device Manager).
- Look for devices with a yellow exclamation mark. That's your first suspect.
- Right-click the device and select 'Properties'.
- Go to the 'Driver' tab and click 'Roll Back Driver' if available. If not, download the latest driver from the manufacturer's site – not Windows Update.
- If rolling back doesn't work, uninstall the driver completely, reboot, and let Windows reinstall it fresh.
Real-world tip: Don't trust 'driver update' tools. They often install the wrong version. Go straight to the hardware vendor's site. For network cards, check Intel or Realtek directly. For graphics, hit AMD or NVIDIA.
2. Corrupt System Files or Memory
If updating drivers didn't stop the bluescreens, the next suspect is memory corruption. The STATUS_INVALID_LDT_SIZE error can fire when Windows tries to allocate memory for the LDT and the memory itself is hosed. I had a client whose laptop would crash every time they opened Chrome. Ran MemTest86 – one stick was failing. Replaced it, and the error never came back.
How to Check Memory
- Press
Windows Key + R, typemdsched.exe, and hit Enter. - Choose 'Restart now and check for problems'.
- Let it run for at least one full pass (takes 20–30 minutes). If it finds errors, replace the bad RAM stick.
System File Check
Sometimes the corruption is in Windows itself. Run these commands in an elevated Command Prompt (right-click Start > Command Prompt Admin):
sfc /scannow
DISM /Online /Cleanup-Image /RestoreHealth
Let SFC finish first. Then run DISM. Reboot after both complete. This fixed the error on a Windows 10 Pro machine that had a corrupted WMI repository.
3. Third-Party Software Crashes
Less common, but I've seen antivirus programs (looking at you, McAfee) and virtualization software like VMware or VirtualBox trigger this. The LDT is used for 16-bit code emulation. If a program tries to set up a bad LDT entry, Windows kills it with this error. One client had an old DOS-based accounting app that would crash every time it tried to print. The app was writing a bad LDT size – updating the app fixed it.
How to Find the Culprit
- Check the Event Viewer (Eventvwr.msc) under Windows Logs > System. Look for the
0xC0000118bug check. It usually lists the offending process. - Uninstall any recently added software, especially security suites or emulators.
- Use Autoruns (Microsoft Sysinternals) to disable non-Microsoft services and see if the error goes away.
Quick test: Boot into Clean Boot mode (msconfig > Services > Hide all Microsoft services > Disable all). Reboot. If the error stops, enable services one by one until you find the bad one.
Quick-Reference Summary Table
| Cause | Fix | Time to Try |
|---|---|---|
| Bad driver | Roll back or update from manufacturer site | 15 minutes |
| Memory corruption | Run mdsched.exe or MemTest86 | 30 minutes to 1 hour |
| Third-party software | Uninstall suspicious apps, check Event Viewer | 20 minutes |
Start with the driver fix – it's the quickest and most common. If that doesn't work, run the memory test overnight. And if you're still seeing the error after both, you're likely dealing with software conflict. Either way, you've got a clear path now.