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

#201
unwind-protect
A Computer GuyHere we go again...
This guy needs to re-test with a pre-3878 Windows 11.

There is no guarantee that this new error is related to the patch.
Posted on Reply
#202
A Computer Guy
RJARRRPCGPThis looks like possibly a Windows PCI-E subsystem fault! :wtf:

It looks extremely sus! It may be due to maximum utilization of PCI-E.
Signal integrity failure?
Posted on Reply
#203
RJARRRPCGP
Update: The SSD is accessible with a power cycle. Does the F1 24 benchmark even do a lot of large file writes?
Posted on Reply
#204
unamed
I don't really believe it's the SSD or the controller used in the SDD at fault.
I have an old Wyse 3290 running Linux and no matter what mSATA device I will install in it, sooner or later it will "hang" the mSATA port (the hang is totaly random, can't even be replicated (can happen in 4h or can happy run for 2+ days doing the same thing); and all power saver is disabled), but it's just the mSATA port that vanishes, everything else is happy running, a power cycle will get the SSD happy running. To make it worst the logs don't really show anything interesting, it doesn't help me that the OS became aware that the mSATA device is no longer there...

Data corruption... Well if the OS is running on the SSD that vanishes and it just happen that the SSD vanishes when the OS wants to write some critical stuff on the SSD (and Windows does a bounch of stuff ("optimization") behind the scene (why would the user need to know that some files are being moved, processed... i don't know maybe because the user is doing something that can easily cause an unexpected reboot...)) then ending up with data corruption is expected. But well it's more if writes happens on the SSD when it vanishes you will have high chances to have data corruption there.

It would be nice to know if the NVMe SSD experiencing the problem is in the slot connected to CPU or MB chipset and the particular CPU or MB chipset. This can easily not be an SSD problem to begin with, the problem can easily be somewhere else, and testing an SSD that is experiencing this problem in a configuration may easily not trigger the problem saw in another configuration. So it's possible that the tests done by MS and Phison are right, in a way, cause it might not be the SSD at fault to begin with... and they just tested for SSD problem in configurations (cpu and/or mb) that just doesn't trigger it...

I don't say that this is not an issue with the ssd controllers, I just say that it's possible to be something else and to replicate it you may need to do it on particular configurations that are triggering it. It won't be a first when there is some software/hardware incompatibility or an update creating such an incompatibility...

You can even have a case when person X working for a particular code was replaced by person Y and person X knew that some stuff will bring more performance but it can cause instability in some case but never let a comment in the code, just had it wrote in his/her notebook and when he/she left he/she took his notebook and person Y happy implement that because it will bring more performance not being aware of the instability part...

L.E.: Also JayzTwoCents can't actually say there what is the problem, for me it looks like some hw register ended up having an invalid value and it needs to be power cycled to get back to a normal (default) value. Main problem is where is that register located, SSD controller, MB chipset, CPU? There is no attempt to use that SSD in another configuration (different CPU manufacturer) to see if he can actually replicate it there also... Network SoC world is full with SoCs that need all type of weird stuff done to have them not hang for all kinda of weird reasons all thanks to hardware bugs...
Posted on Reply
#205
RainingTacco
If microsoft is pushing new operating modes and features that engage SSD in a way that can cause failures in some SSDs, then it's a duty of microsoft to put appropriate safeguards and contact SSD industry liasons to discuss the issue. As of now, i blame 100% microsoft for this.
Posted on Reply
#206
trparky
I reinstalled the offending update along with the new preview update that's available on two of my machines, my desktop and my brother's notebook. Both of which aren't running SSDs that have Phison controllers. I then did a bit of a torture test on the systems.

With the help of a Powershell script, I have repeatedly written over 64 GBs of data to my SSDs as a sort of torture test along with a test with Crystal Disk Mark. Both systems still have functioning SSDs after this torture test.

To test your own system, rename the file to a file with ".ps1" as the extension and run it in a Powershell environment. It accepts one argument, -DriveLetter. It's not required though, it will run by default on the C drive.
Posted on Reply
#207
DaemonForce
What are the OS builds and updates? This literally tells us nothing and I am still seeing creators lose their shit over problems related to this SSD catastrophe.
Messages like 鯖に書いてる手順の簡易版を書いときますね。 設定→Windows Update→更新の履歴→更新プログラム 履歴にある「KB5062660」「KB5063878」があればアンインストール&再起 です。
So clearly it's a problem for the normal parts of the world that never reinstall, move hardware of any kind or do anything outside of general PC use.
Posted on Reply
#208
remixedcat
trparkyI reinstalled the offending update along with the new preview update that's available on two of my machines, my desktop and my brother's notebook. Both of which aren't running SSDs that have Phison controllers. I then did a bit of a torture test on the systems.

With the help of a Powershell script, I have repeatedly written over 64 GBs of data to my SSDs as a sort of torture test along with a test with Crystal Disk Mark. Both systems still have functioning SSDs after this torture test.

To test your own system, rename the file to a file with ".ps1" as the extension and run it in a Powershell environment. It accepts one argument, -DriveLetter. It's not required though, it will run by default on the C drive.
And do it on phison based drives
Posted on Reply
#209
trparky
remixedcatAnd do it on phison based drives
I don't have any at my disposal.

My desktop has a Samsung 970 EVO and a Samsung 980 while my brother's notebook has a Micron 2400 and a Western Digital Black SN770.
Posted on Reply
#210
remixedcat
trparkyI don't have any at my disposal.

My desktop has a Samsung 970 EVO and a Samsung 980 while my brother's notebook has a Micron 2400 and a Western Digital Black SN770.
And I'm not gonna risk my brand new firecuda on that mess either lol
Posted on Reply
#211
trparky
The next torture test is going to be on my cheap Acer notebook that by the little resarch I've done has a SKHynix SSD, a HFM512GD3JX016N to be specific. I'm updating it now and will be back in about an hour with some results. Wish me luck.

So far, so good with the Acer notebook. I installed the problematic update along with this month's preview update along with a number of other software updates including the biggest one of them all, an update to Microsoft Visual Studio. I've ran my Powershell script twice, writing 140 GBs of random data to the drive. No issues—no BSOD, no slow downs, etc.
Posted on Reply
#212
remixedcat
trparkyThe next torture test is going to be on my cheap Acer notebook that by the little resarch I've done has a SKHynix SSD, a HFM512GD3JX016N to be specific. I'm updating it now and will be back in about an hour with some results. Wish me luck.

So far, so good with the Acer notebook. I installed the problematic update along with this month's preview update along with a number of other software updates including the biggest one of them all, an update to Microsoft Visual Studio. I've ran my Powershell script twice, writing 140 GBs of random data to the drive. No issues—no BSOD, no slow downs, etc.
A can't find the controller on that one...
Posted on Reply
#213
A Computer Guy
So I tried another test today with 24H2 and the offending update this time on native hardware 4750G on Asrock X470 Gaming ITX/ac (PCIe 3.0) using an old Crucial P1 1TBdrive I had (Silicon Motion SM2263 PCIe 3.0 x4 NVME v1.3) HWiNFO64 says the drive does not support HMB. Coping around 57GB file full of Linux ISO's and zipping and unzipping. Didn't get any errors. Drive peaked at 75c.
Posted on Reply
#214
remixedcat
A Computer GuySo I tried another test today with 24H2 and the offending update this time on native hardware 4750G on Asrock X470 Gaming ITX/ac (PCIe 3.0) using an old Crucial P1 1TB drive I had (Silicon Motion SM2263 PCIe 3.0 x4 NVME v1.3) HWiNFO64 says the drive does not support HMB. Coping around 57GB file full of Linux ISO's and zipping and unzipping. Didn't get any errors. Drive peaked at 75c.
that's sorta promising for my sk hynix wich uses a
SM2269XT Controller
but I'm not aboutta try my luck tho I dont wanna have to get another nvme and I need my spare cash for an exos or wd ultrastar next month.
Posted on Reply
#215
trparky
I took another look at JaysTwoCent's video and what makes me really scratch my head is that he said the crash happened not during what people typically said causes the SSD to crash and that it happened during a game benchmark.

That seems to point to something (at least in my mind) happening on the PCIe bus itself since it happened when using the GPU heavily. I'm almost wondering if some kind of data corruption issue is happening when operating in either PCIe version 4 or higher.
Posted on Reply
#216
chrcoluk
A Computer GuyHere we go again...
My main issue with Jayz's video is should have done at least one of two things.

Either.

1 - Test the SSD again on a install of windows that has never had the update installed, yes more work for the video but would have strengthened his case, if the drive no longer had the symptoms, its stronger evidence.
Or
2 - If he has multiple units of the same drive, test those as well, to rule out just a faulty unit which he might have.

Reason being I had similar on my 850 Pro on windows 10 some years ago. The video for me isnt conclusive.
Posted on Reply
#218
Sora
unamedI don't really believe it's the SSD or the controller used in the SDD at fault.
I have an old Wyse 3290 running Linux and no matter what mSATA device I will install in it, sooner or later it will "hang" the mSATA port (the hang is totaly random, can't even be replicated (can happen in 4h or can happy run for 2+ days doing the same thing); and all power saver is disabled), but it's just the mSATA port that vanishes, everything else is happy running, a power cycle will get the SSD happy running. To make it worst the logs don't really show anything interesting, it doesn't help me that the OS became aware that the mSATA device is no longer there...

Data corruption... Well if the OS is running on the SSD that vanishes and it just happen that the SSD vanishes when the OS wants to write some critical stuff on the SSD (and Windows does a bounch of stuff ("optimization") behind the scene (why would the user need to know that some files are being moved, processed... i don't know maybe because the user is doing something that can easily cause an unexpected reboot...)) then ending up with data corruption is expected. But well it's more if writes happens on the SSD when it vanishes you will have high chances to have data corruption there.

It would be nice to know if the NVMe SSD experiencing the problem is in the slot connected to CPU or MB chipset and the particular CPU or MB chipset. This can easily not be an SSD problem to begin with, the problem can easily be somewhere else, and testing an SSD that is experiencing this problem in a configuration may easily not trigger the problem saw in another configuration. So it's possible that the tests done by MS and Phison are right, in a way, cause it might not be the SSD at fault to begin with... and they just tested for SSD problem in configurations (cpu and/or mb) that just doesn't trigger it...

I don't say that this is not an issue with the ssd controllers, I just say that it's possible to be something else and to replicate it you may need to do it on particular configurations that are triggering it. It won't be a first when there is some software/hardware incompatibility or an update creating such an incompatibility...

You can even have a case when person X working for a particular code was replaced by person Y and person X knew that some stuff will bring more performance but it can cause instability in some case but never let a comment in the code, just had it wrote in his/her notebook and when he/she left he/she took his notebook and person Y happy implement that because it will bring more performance not being aware of the instability part...

L.E.: Also JayzTwoCents can't actually say there what is the problem, for me it looks like some hw register ended up having an invalid value and it needs to be power cycled to get back to a normal (default) value. Main problem is where is that register located, SSD controller, MB chipset, CPU? There is no attempt to use that SSD in another configuration (different CPU manufacturer) to see if he can actually replicate it there also... Network SoC world is full with SoCs that need all type of weird stuff done to have them not hang for all kinda of weird reasons all thanks to hardware bugs...
jay is slapping an uncooled nvme in a slot and calling it a day, the abused T500 does not come with a heatsink in the sku Jaz used.
Posted on Reply
#219
A Computer Guy
chrcolukMy main issue with Jayz's video is should have done at least one of two things.

Either.

1 - Test the SSD again on a install of windows that has never had the update installed, yes more work for the video but would have strengthened his case, if the drive no longer had the symptoms, its stronger evidence.
Or
2 - If he has multiple units of the same drive, test those as well, to rule out just a faulty unit which he might have.

Reason being I had similar on my 850 Pro on windows 10 some years ago. The video for me isnt conclusive.
I agree 100%. He did his viewers dirty by not doing that. He could have even slapped a Windows 10 on it rather quickly and re-tested for the same issue but no. It's just too simple a task these days to see simple troubleshooting and process of elimination to help narrow down to the root cause. Sigh. :slap:
Posted on Reply
#220
eyenix
Hi guys, do you have any news on this? I have a second Silicon Power UD90 SSD with a Phison E21 controller that has been reported by several users and acquaintances as having problems and some even had their SSD broken.
Posted on Reply
#221
unwind-protect
So has there been any confirmation at all after the original Japanese complaints?
Posted on Reply
#224
DaemonForce
結果
•イベントビューアに disk 153 / 129 / 55 が発生 ボリュームラベルバグり 写真2
•CrystalDiskInfo の SMART「電源投入回数」が爆伸び(明らかに異常)。
•「おお、これは壊れる寸前やん✌︎(‘ω’✌︎ )」って状況までは再現できた。
ただし実際には残り1%容量まで削ってもSSDは壊れず耐えた。
「鬼畜で攻めればエラーは出せる。でもトドメまではなかなか難しい」って印象。

Result
• disk 153/129/55 occurs in event viewer Volume label bug Photo 2
•CrystalDiskInfo's SMART "number of power-ups" has increased dramatically (obviously abnormal).
•I was able to recreate the situation by saying, "Oh, this thing is on the verge of breaking ✌ ︎(‘ω’ ✌︎ )".
However, in reality, the SSD survived without breaking even when the remaining 1% capacity was reduced.
"If you attack with a devil, you can get an error. But it's quite difficult to get to the final blow.''.

Magical.
Posted on Reply
#225
trparky
I read the translation of the Japanese text and it made no sense to me. It read like a badly AI-translated version.
Posted on Reply
Add your own comment
Nov 19th, 2026 14:49 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts