Monday, August 18th 2025

Microsoft Windows 11 24H2 Update May Cause SSD Failures

Microsoft issued an OS security update as part of the August 2025 Patch Tuesday cycle to counter the Lumma stealer and similar malware and exploits. The Windows 11 24H2 cumulative package KB5063878 was included in that rollout. Early deployments returned an installation error, 0x80240069, which Microsoft acknowledged and later addressed. Community testing has since identified a reproducible pattern in which certain SSDs, when running sustained sequential writes of approximately 50 GB, become inaccessible and display unreadable SMART telemetry. Follow-up analysis and vendor feedback to date primarily point to solid-state drives, although conclusions about traditional spinning hard disk drives remain unconfirmed.

A Japanese tester NichePCGamer published a 21-drive matrix and a repeatable procedure: copy a large game folder, write an approximately 62 GB archive composed of many video files to the target SSD, then decompress that archive on the same device. Failures were reported most frequently when controller utilization exceeded 60% during continuous writes. Reported devices include several Phison-based NVMe models and other drives using different controllers. Symptoms reportedly recur after a reboot, which often restores visibility temporarily but does not eliminate the underlying fault, creating a risk of file corruption if the same workload is repeated. Microsoft and SSD vendors are investigating and have not yet published a definitive root-cause report. Until vendors and Microsoft publish formal guidance, users with SSDs should back up important data, avoid long sequential transfers on systems updated to KB5063878, and monitor vendor and Microsoft advisories for SSD-specific firmware, driver, or OS fixes.

Update 14:02 UTC: Phison has posted an update, acknowledging the issue.

Update Aug 21st: Phison has reached out to us to let us know that some documents about this issue are fake.
Sources: via Neowin, Notebookcheck
Add your own comment

278 Comments on Microsoft Windows 11 24H2 Update May Cause SSD Failures

#226
DaemonForce
Uh, no. That's been the case with EN-JP translation forever. These languages are not best friends.
Posted on Reply
#227
KAIHAN
It's full of slang and even Japanese people don't understand it very well.
I heard that this solved the problem, so I posted this, but looking at it now, it seems that it still hasn't been solved. I'm sorry.
However, it seems that you're also having trouble reproducing the problem, so I don't think you need to worry about it.

In this test, there were no problems in a clean install environment, regardless of whether KB5063878 was installed or not.

chimolog.co/ssd-vs-kb5063878/
Posted on Reply
#228
_roman_
In short there are the next issues with Windows 11 pro. Waiting and pausing the updates may be very smart now.
Posted on Reply
#229
pf100
Throughout all of this drama I've stayed on Windows 11 Enterprise 23H2 and kept it on 23H2 with GRC's InControl. When 24H2 was released I tried it for a couple of days and quickly realized that it had serious problems. 23H2 is supported on Enterprise versions of windows until Nov 10, 2026 (for another 14 months as I type this) which is exactly how long I'll be running it.
Posted on Reply
#230
trparky
24H2, minus this issue, of course, is pretty damn stable. I upgraded a system in my house that was running 23H2 to 24H2 and afterwards the system was running a whole lot faster, more fluid.
Posted on Reply
#231
jnv11
JayzTwoCents released a new video showing that a BIOS update killed the bug that caused his SSD to fail. His SSD is now working. He thinks that an AGESA bug might be the root cause:

Posted on Reply
#232
unamed
jnv11JayzTwoCents released a new video showing that a BIOS update killed the bug that caused his SSD to fail. His SSD is now working. He thinks that an AGESA bug might be the root cause:

I can't take him serious to be honest. He ignores so many technical posibilities in his videos...
For the SSD problem the fact that the system was stable before doesn't mean that he didn't had some particular component failing just when he made that update, for example some cap could had went bad on that SSD and while the SSD works fine in most cases in some particular rare cases when that cap should actually do its job he just has the failure or he always had a lose connector there and in time it degrated to the the point that the SSD is no longer getting the proper voltage and well while initially it worked in time components degrated to the point it's no longer stable with that lower voltage, also the mb might be faulty to begin with and push lower voltage into that ssd...
(Heck i had a pc that worked fine for 4 years in a row without anything being touched inside it or the pc being moved and one day it just started to no longer boot anything and it turned out to be 1 dimm not staying properly in the slot so the argument that it worked fine before the update and not after so it must the the update is just silly.)

Again I don't say it's not the update, I just say that saying that it's 100% the update while there can be other reasons it's just bad.

He even admits of using a beta Bios on a motherboard made by a manufacturer that well it's know for having mb that in some cases are malfunctioning with the cpu (to say it nicely here and not to say destroying the cpu). And in this case he might just see a prefailure behaviour of the cpu...

The are so many variables that he is not taking into consideration or he just dismissed them without real arguments that makes him no better in reasoning than some flat earth believers that refuse to understand that with their models u should be able to see some high peaks but u don't.

And the behaviour he is experiencing later when he is actually putting the SSD back into the test machine (where ports and connectors will wear faster than in a normal pc) is more of a bad SSD or bad connection in the port than a Windows update failure.

Also he is just... when asking if someone on Intel is experiencing it. He is so poor that he can't afford to build a cheap Intel system and test it himself?!?

I'm not attacking him at all, just poiting to how badly he is failing in this video. He say in this video about the journalist not doing the homework properly but in this video he also didn't do his homework properly...

He obvious doesn't have experience with what is inside the bios for a mb or gpu, else he would had knew how buggy some of them can be, while the mb or gpu actually works with that buggy bios...

Wonder if he is expanding that theory regarding bios updates "if it's not broken why fix it" to other things like android updates, software updates because well it works so why would u need update. He should be consistent and well disable windows updates for the same reason as he is not updating the bios because well same as a bios update an os update can cause him not get consistent testing results, he should do the same with all the software running on the machine used for testing actually he should fully disable any security software because that might be updated and make him get inconsistent results. I'm not serious here, just using his own words to prove that he is not consistent with the reasoning.

L.E.: if he was so sure that it is windows causing the problem and he is so poor that he can't afford to build a cheap intel system just to test the SSD there then well why he didn't tried to test this sdd in Linux?!? I would had loved to see a dmesg log after the SSD vanished, just a reset (not a power off / power on) and boot from a live usb stick and a dmesg log. In term of actually trying to debug the problem he has done nothing: no attempt to boot in another OS after the SSD vanished, no attempt to run the SSD in another machine to try to replicate it. After the bios update all he knows is that now he can't trigger it and he is assuming it's the agesa update when it can be some voltages/power adjustment done in the new bios that fixed it (something that asrock might had actually done in the attempt to prevent the malfunctioning with particular cpus).

Also i fail to understand why he didn't updated that bios to the latest stable. It makes no sense to just upgrade it to next version and just stop there.
Posted on Reply
#234
unwind-protect
I have FAT CF cards (no exFAT) with 64 and 128 GB in my camera. Bogus video.
Posted on Reply
#235
pf100
unwind-protectI have FAT CF cards (no exFAT) with 64 and 128 GB in my camera. Bogus video.
Not really. It depends on the sector size and what an OS supports and what the partition is used for. If you use 1k sectors a fat32 partition can be 16TB but Windows won't read it natively. Until recently Windows only supported 32GB fat32 partitions. What he's saying in the video is that for general use a 32 GB partition is the maximum reasonable size for a FAT32 partition to minimize wasted disk usage with mixed file sizes because of excessive slack space. The reason that doesn't matter on a camera so much is that videos and large picture files use up most of the slack space so it's fine. If you put only tiny files on a 128GB fat32 partition it would fill up the drive before it ran out of space because of all the wasted slack space. That's why Windows stuck with 32GB partitions for so long. Drives are so much bigger now that wasted space isn't as big of a deal as it used to be.
Posted on Reply
#236
unwind-protect
pf100Not really. It depends on the sector size and what an OS supports and what the partition is used for. If you use 1k sectors a fat32 partition can be 16TB but Windows won't read it natively. Until recently Windows only supported 32GB fat32 partitions. What he's saying in the video is that for general use a 32 GB partition is the maximum reasonable size for a FAT32 partition to minimize wasted disk usage with mixed file sizes because of excessive slack space. The reason that doesn't matter on a camera so much is that videos and large picture files use up most of the slack space so it's fine. If you put only tiny files on a 128GB fat32 partition it would fill up the drive before it ran out of space because of all the wasted slack space. That's why Windows stuck with 32GB partitions for so long. Drives are so much bigger now that wasted space isn't as big of a deal as it used to be.
The CF cards I am talking about are formatted by Canon cameras and read fine on Linux, macOS and FreeBSD. Dunno about Windows. Linux and FreeBSD can also create them with a variety of tools, but letting the camera do it is more straightforward.
Posted on Reply
#237
remixedcat
A Computer GuyPrerelease firmware an issue? Timestamp 3:49
its not the drives or firmware and I'm talking to one of the companies that makes ssds and they do not ship beta or engineering firmware with their drives and some of their drives were listed as affected.


Posted on Reply
#238
pf100
unwind-protectThe CF cards I am talking about are formatted by Canon cameras and read fine on Linux, macOS and FreeBSD. Dunno about Windows. Linux and FreeBSD can also create them with a variety of tools, but letting the camera do it is more straightforward.
You missed my point which was slack space. Because of slack space, your camera's 128GB FAT32 partition was fine (because it only used large files which minimized wasted slack space) when a 32GB FAT32 partition made more sense on Windows. The video was not bogus.

Maybe this video will help explain why 32GB was the biggest FAT32 partition size you wanted on Windows *at the time*, and why a 128GB FAT32 partitrion was fine with your camera. Remember that small files wastes half of the space on your 128GB CF card but large files (video) won't.

Posted on Reply
#239
Sora
IF you're getting this on a 990 pro, samsung now has released 7B2QJXD7.
Posted on Reply
#240
_roman_
pf100Not really. It depends on the sector size and what an OS supports and what the partition is used for. If you use 1k sectors a fat32 partition can be 16TB but Windows won't read it natively. Until recently Windows only supported 32GB fat32 partitions.
Windows 10 pro and 11 pro can not even properly format USB drives. From the Windows standpoint FAT32 ends at 4GB or 2GB. 16GB USB drives are formatted wrong.

128GB USB drives are also wrong formatted in windows 11 pro with ntfs.

My workaround is mkfs.???? /??? (i translate the binary to make filesystem =mkfs) for e.g. 16GB / 4GB fat32 usb sticks or 128GB NTFS sticks.

Microsoft should fix the format dialog background software stack or fix the stuff shown in the format dialog. They are both broken. EXFAT is not FAT32 - it can not be read my my devices.
~2 Weeks or less, w11 pro refurbisher license, all updates done: Two partitions are not a single NTFS partition in the size of 128GB for my USB transfer stick.
Stuff formated with e.g. www.man7.org/linux/man-pages/man8/mkfs.vfat.8.html or the other filesystem familes are all readable in w10 pro, w11 pro or any other device so far. Definitely a windows 10 pro and 11 pro issue with the bad build in format dialog.

Seems wikipedia but no other valid source for max file system limits: en.wikipedia.org/wiki/File_Allocation_Table#Types

Edit: I will use my ignore list and not engange. The issue was explained for someone who knows the core basics of computers. Which are still the same since windows 3.11 afaik / linux kernel 2.0.0.
Posted on Reply
#241
Sora
nothing is broken in windows in regards to the format utility
Posted on Reply
#242
remixedcat
Ok I confirmed w 3 more manufacturers they do not ship ssds with beta/engineering firmware.

Crucial


Seagate


Sandisk/WD
unwind-protectI have FAT CF cards (no exFAT) with 64 and 128 GB in my camera. Bogus video.
Dave Plummer made millions of dollars scamming people
Posted on Reply
#243
pf100
_roman_Windows 10 pro and 11 pro can not even properly format USB drives. From the Windows standpoint FAT32 ends at 4GB or 2GB. 16GB USB drives are formatted wrong.

128GB USB drives are also wrong formatted in windows 11 pro with ntfs.

My workaround is mkfs.???? /??? (i translate the binary to make filesystem =mkfs) for e.g. 16GB / 4GB fat32 usb sticks or 128GB NTFS sticks.

Microsoft should fix the format dialog background software stack or fix the stuff shown in the format dialog. They are both broken. EXFAT is not FAT32 - it can not be read my my devices.
~2 Weeks or less, w11 pro refurbisher license, all updates done: Two partitions are not a single NTFS partition in the size of 128GB for my USB transfer stick.
Stuff formated with e.g. www.man7.org/linux/man-pages/man8/mkfs.vfat.8.html or the other filesystem familes are all readable in w10 pro, w11 pro or any other device so far. Definitely a windows 10 pro and 11 pro issue with the bad build in format dialog.

Seems wikipedia but no other valid source for max file system limits: en.wikipedia.org/wiki/File_Allocation_Table#Types

Edit: I will use my ignore list and not engange. The issue was explained for someone who knows the core basics of computers. Which are still the same since windows 3.11 afaik / linux kernel 2.0.0.
What are you even talking about? I explained slack space but you insist on denying that it exists and it makes you mad, then you start talking about unrelated things? You need to learn the importance of minimizing slack space. I am aware that I am ignored so keep denying that slack space exists, I guess.
remixedcatOk I confirmed w 3 more manufacturers they do not ship ssds with beta/engineering firmware.
Dave Plummer made millions of dollars scamming people
Are you sure that he's scamming people? Are you sure that lots of people didn't get beta firmare? No, you're not sure. You're going by what a corporation told you. I'm not saying that he did or didn't scam people, I'm saying you're making accusations without proof.
Posted on Reply
#244
remixedcat
pf100What are you even talking about? I explained slack space but you insist on denying that it exists and it makes you mad, then you start talking about unrelated things? You need to learn the importance of minimizing slack space. I am aware that I am ignored so keep denying that slack space exists, I guess.



Are you sure that he's scamming people? Are you sure that lots of people didn't get beta firmare? No, you're not sure. You're going by what a corporation told you. I'm not saying that he did or didn't scam people, I'm saying you're making accusations without proof.
Did you watch the video??

And I trust ssd manufacturers way more than I would ever trust microsoft.

Who in their right mind would release a product with beta firmware and risk their whole brand for what?? No decent company would.

I wish ppl would stop standing up for one of the most vile companies in existence
Posted on Reply
#245
chrcoluk
The slack space is largely moot these days, especially on modern file systems. First that the average file size has shot up, but also that the smaller the cluster size, the larger the metadata overhead.

There is a few niche cases where it may matter such as maildir email storage, or something like a git which has a ton of tiny sized files. But for your typical end user it's moot.

Even more so on larger drives, as using small clusters the overhead starts to become extreme affecting performance. A 4tb 4k ntfs has nearly a billion records to manage.

My odd performing SA510 throughput tripled (and i/o delay) moving from 4k to 16k clusters, the overhead on 4k was really harming that drive. It went from 560/150 reads (pSLC/native) to 560/450. A few testers have started to suggest 32k as optimal for SSD's and 64k to a couple of megabytes for spindles.
remixedcatDid you watch the video??

And I trust ssd manufacturers way more than I would ever trust microsoft.

Who in their right mind would release a product with beta firmware and risk their whole brand for what?? No decent company would.

I wish ppl would stop standing up for one of the most vile companies in existence
To be fair one company did use experimental features as they wanted an edge in the market, and in the end it destroyed their rep and brand, remember the old OCZ SSDs?
Posted on Reply
#246
remixedcat
chrcolukThe slack space is largely moot these days, especially on modern file systems. First that the average file size has shot up, but also that the smaller the cluster size, the larger the metadata overhead.

There is a few niche cases where it may matter such as maildir email storage, or something like a git which has a ton of tiny sized files. But for your typical end user it's moot.

Even more so on larger drives, as using small clusters the overhead starts to become extreme affecting performance. A 4tb 4k ntfs has nearly a billion records to manage.

My odd performing SA510 throughput tripled (and i/o delay) moving from 4k to 16k clusters, the overhead on 4k was really harming that drive. It went from 560/150 reads (pSLC/native) to 560/450.


To be fair one company did use experimental features as they wanted an edge in the market, and in the end it destroyed their rep and brand, remember the old OCZ SSDs?
Well the rest have learned from that at least. I haven't heard many others doing that since... only ones I did hear of were shady no name tofu dreg brands..
Posted on Reply
#247
_roman_
pf100What are you even talking about?
I explained that windows 10 pro and 11 pro has a broken "format dialog", even recently. You may check the manpage for mkfs. Anyway - I think i explained it quite well. I do not need to learn anything. Topic over. Some issues are realted to that. Even for myself. I do expect more for a 150€ operating system.

I assume this is the right box for windows 11 pro.
geizhals.at/microsoft-windows-11-pro-64bit-deutsch-pc-fqc-10534-a2611704.html

Edit: Windows 10 home oem / pro did format usb sticks correctly with ntfs several years ago. Formatting bigger usb sticks with fat is broken for a very long time.
Posted on Reply
#248
A Computer Guy
remixedcatWho in their right mind would release a product with beta firmware and risk their whole brand for what?? No decent company would.
They might do so for reviewers for early access reviews who then later sell off those drives because they don't need them anymore. However I also think the overall quantity of those kinds of drive releases would be so statistically insignificant you would never see issue complaints in mass as we have see with this problem.

Since it seems nobody has any answers for the problem I'm inclined to believe the scenario for the problem is a very corner case scenario or it's just way overblown and people are learning consumer based flash just isn't as reliable as everyone thinks it is. ( I haven't kept up with current events in the past weeks so I might be missing something. )
SoraIF you're getting this on a 990 pro, samsung now has released 7B2QJXD7.
I've installed this update and so far so good but I'm also running Windows 10 so I wasn't having the issue anyway.
semiconductor.samsung.com/consumer-storage/support/tools/
*(7B2QJXD7) To address the intermittent non-recognition and blue screen issue. (Release: September 2025)
Posted on Reply
#249
remixedcat
A Computer GuyThey might do so for reviewers for early access reviews who then later sell off those drives because they don't need them anymore. However I also think the overall quantity of those kinds of drive releases would be so statistically insignificant you would never see issue complaints in mass as we have see with this problem.
And I saw comments even saying "bought brand new from newegg and...." or "just got this from Amazon 2 months ago and...." kinda comments to so those are retail drives...

adding this too from my screenshot archives (ssd_data phison rep deleted their reddit account sadly)

phison would be pissed if any partners did this with their controllers
Posted on Reply
#250
Sora
pf100What are you even talking about? I explained slack space but you insist on denying that it exists and it makes you mad, then you start talking about unrelated things? You need to learn the importance of minimizing slack space. I am aware that I am ignored so keep denying that slack space exists, I guess.



Are you sure that he's scamming people? Are you sure that lots of people didn't get beta firmare? No, you're not sure. You're going by what a corporation told you. I'm not saying that he did or didn't scam people, I'm saying you're making accusations without proof.
Dave was sued for it, so we know its true.
www.atg.wa.gov/news/news-releases/attorney-general-s-office-sues-settles-washington-based-softwareonlinecom
_roman_I explained that windows 10 pro and 11 pro has a broken "format dialog", even recently. You may check the manpage for mkfs. Anyway - I think i explained it quite well. I do not need to learn anything. Topic over. Some issues are realted to that. Even for myself. I do expect more for a 150€ operating system.

I assume this is the right box for windows 11 pro.
geizhals.at/microsoft-windows-11-pro-64bit-deutsch-pc-fqc-10534-a2611704.html

Edit: Windows 10 home oem / pro did format usb sticks correctly with ntfs several years ago. Formatting bigger usb sticks with fat is broken for a very long time.
You keep saying broken, but you're wrong.
The Format GUI intentionally limits itself to ntfs and exFAT because of the expectations of what high capacity removable drives are to be used for.

They could always be formatted with third party UI's (using the same format api) and command line tools to FAT32.
Posted on Reply
Add your own comment
Nov 19th, 2026 14:59 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts