Error Messages

WHEA Uncorrectable Error (0x124): What It Means and How to Fix It

Updated October 2, 2026 •7 min read

Your PC crashes to a blue screen with WHEA_UNCORRECTABLE_ERROR. It might happen while gaming, under heavy load, or seemingly at random. The WHEA uncorrectable error sounds alarming, and it is a hardware problem, but it’s often not the drive. Here’s how to narrow it down in a sensible order, and how to tell when your data really is at risk.

What the WHEA uncorrectable error means

WHEA stands for Windows Hardware Error Architecture. It’s the part of Windows that receives hardware error reports. Modern processors have a built-in error-reporting system called Machine Check Architecture (MCA). When the CPU detects a hardware fault it can’t recover from, it raises a machine check, and Windows stops with stop code 0x124 (WHEA_UNCORRECTABLE_ERROR).

The error is reported by the CPU, but the fault isn’t necessarily in the CPU. The processor sits in the middle of everything: memory, the PCIe bus, and the devices on it. A problem with any of them can surface as a machine check.

Common causes, roughly most to least likely

Overclocking or voltage tuning. CPU overclocks, undervolts, and memory profiles (XMP/EXPO) that are on the edge of stable are a very common trigger.
Overheating. A clogged cooler, dried-out thermal paste, or a failing fan can push the CPU or VRMs past safe limits.
RAM. A faulty module or one that’s unstable at its rated speed.
Power supply. An aging or undersized PSU may not deliver stable voltage under load.
Outdated BIOS, chipset drivers, or other drivers. BIOS updates often include CPU microcode and stability fixes.
The CPU itself. Less common, but possible.
Storage. A failing NVMe SSD, a poorly seated M.2 drive, or a storage controller problem can produce PCIe errors that end in a WHEA crash.

Read the error before you start swapping parts

Windows logs details about the fault. Open Event Viewer (right-click Start > Event Viewer) > Windows Logs > System, and look for entries from the source WHEA-Logger around the time of the crash. Event ID 18 is a fatal hardware error. Its details name the reported component, for example:

Processor Core or Cache Hierarchy Error → points toward CPU stability: overclock, voltage, heat, or the CPU itself.
Memory → test RAM first.
Bus/Interconnect or a PCI Express device → something on the PCIe bus, which can be a graphics card or an NVMe SSD.

This won’t always pinpoint the part, but it tells you where to start.

Step 1: Protect your data before heavy testing

Most WHEA errors don’t damage stored data. Even so, every crash interrupts whatever was being written, and the tests below put the system under load. If you don’t have a current backup, copy your important files to an external drive or cloud storage now, while the system still runs. See the 3-2-1 backup rule for a setup that keeps this from being urgent next time.

Step 2: Remove overclocks and reset the BIOS

Enter your BIOS/UEFI setup and load optimized defaults. That turns off CPU overclocks, undervolts, and XMP/EXPO memory profiles. Also undo any tuning in Windows apps from your motherboard or GPU maker. If the crashes stop, re-enable settings one at a time, starting with the memory profile, to find the unstable one.

Step 3: Check temperatures

Use a monitoring tool such as HWiNFO to watch CPU temperatures at idle and under load. If temperatures climb very high under load, clean the dust out of the heatsink and fans, make sure all fans spin, and consider fresh thermal paste. Laptops with blocked vents are especially prone to this.

Step 4: Update the BIOS and drivers

Download the latest BIOS, chipset drivers, and, if you have an NVMe SSD, its firmware from your motherboard, PC, or SSD maker’s support site. Follow their instructions carefully. A BIOS update interrupted by a crash or power loss can leave a board unbootable, so do it with defaults loaded and the system stable.

Step 5: Test the RAM

Press Win + R, type mdsched.exe, and choose Restart now and check for problems. For a deeper test, run a bootable tool like MemTest86 for several passes. If you have more than one module, testing them one at a time helps isolate a bad stick.

Step 6: Consider the power supply and the CPU

If crashes happen under load (gaming, rendering) and everything above checks out, the PSU is a common culprit, especially if it’s old or close to its rated limit. Swapping in a known-good PSU is the cleanest test. A faulty CPU is usually the last thing left after everything else has been ruled out.

Suspect the drive if:

Event Viewer shows WHEA-Logger entries naming a PCI Express device that corresponds to your NVMe SSD.
Crashes cluster around disk-heavy activity: large copies, game installs, backups.
The SSD disappears from the BIOS or from Windows after a crash, then comes back.
SMART data shows problems. In CrystalDiskInfo or a similar tool, check Critical Warning, Media and Data Integrity Errors, and Available Spare on NVMe drives.
You also see storage-specific errors such as KERNEL_DATA_INPAGE_ERROR or file-copy errors.

If that’s your situation:

Reseat the SSD. Power off, unplug, remove the M.2 drive, and reseat it firmly. If your board has another M.2 slot, try that one.
Copy your most important files first, to a different drive, before any scans or firmware updates.
Avoid full-surface scans and repeated chkdsk /r runs on a drive that’s throwing errors.
If the SSD isn’t detected at all, see SSD not detected. Many SSDs scramble or encrypt data internally, so recovery usually depends on getting the original controller working again. Repeated power cycling doesn’t help.

The bottom line

A WHEA uncorrectable error is a hardware fault reported by the CPU’s error-checking system. The usual suspects are overclocks, heat, RAM, power, and outdated firmware, not the drive. Work through them in that order, and use Event Viewer’s WHEA-Logger entries to point you in the right direction. If the evidence points to your SSD, stop testing and get your data off first.

If a crashing PC has left you unable to reach important files, our desktop and PC data recovery service recovers data from failed drives for one flat $1,200 price for a single device. Shipping is free and prepaid, and you pay nothing unless we recover your data.