18 comments

[ 3.0 ms ] story [ 42.4 ms ] thread
Reading the article, the headline is inaccurate/wrong: the problem is crappy SSDs (an age-old problem) that can't handle/brick under some workload presumably involving high write rates. This Windows update happens to include an application that can generate such a workload.
Neowin. My old account is from 2001. Although I stopped reading around 2010 when I found them too Anti-Apple and extremely Pro Microsoft. But there seems to be a surge in submission and I am not sure if it is good enough for HN.
>The issue purportedly surfaces during heavy write operations to certain NVMe SSDs as well as HDDs, especially when continuous sustained writes approach 50 GB on drives and exceed 60 per cent controller usage.
yep bricked my beelink SER9 which I had for 8 months only

and i see the same error under windows update on my backup pc now:

2025-08 Cumulative Update for Windows 11 Version 24H2 for x64-based Systems (KB5063878) (26100.4946) Failed to install on ‎17/‎08/‎2025 - 0x80073712

articles says August but I had this update stuck on failed for 2 months. bricked last pc now waiting for it to brick my current one.

how long does it take MS to fix something..

Again? I swear this has happened before
Another real reason not to move to 11. Thank you Microsoft.
Just a reminder to switch to Linux in October when win10 reaches EOL.
The most basic thing an operating system should do is keep your files and information safe. Microsoft Windows has been steadily declining in that direction. It's time to punish them and let them see the consequences. It's time to switch to Linux Debian or Ubuntu.
Is this actually breaking the drives or just making particular files on that particular partition un-readable by Windows? Stated another way: Could a system that dual-boots Linux and Windows actually brick the drives system-wide by what Windows is doing?
Great. So this might be the cause of ally problems. I just happen to cloning drives to swap out with a larger one and Windows 11 is so slow. Also seems there were issues found that needing fixing after cloning.
(comment deleted)
What actions can we take here? Is there a list of the particular drives that would be affected by this?
Apparantly a WD_BLACK SN770 1TB, fw 731100WD, drive is affected by this update. I can confirm that the update to 24H2 was started and that the SSD has gotten a corrupted partition table. Diskpart shows nothing, TestDisk shows double entries and not markers and the log says check_part_gpt failed for partition. With the corresponding line No FAT, NTFS, ext2, JFS, Reiser, cramfs, or XFS marker in the console. I am currently rebuilding the partition table from gpt backup, but I don't know if the 24H2 update has a repartitioning action in it. Also the boot partition is Bitlocker enabled. It is a nice challenge to try and recover it fully.

But, if the update hasn't been completed and the hammering continues, then I think chances are it might happen again.

My guess is that the firmware of the ssd needs to be updated first to account for the HMD bug?

Oh well...

If your rig is stable and dont care about features, block all updates and turn the service off. If you have a prod. rig that is stable, leave it to MS to frack it up. /s