Windows 10 update KB5034441 is still broken, more than a month after release

M76

M76

[H]F Junkie
Joined
Jun 12, 2012
Messages
14,039
wup.PNG


During the January update round MS issued a security update for Windows 10 that should fix a security issue which could allow an attacker to bypass Bitlocker encryption using the Windows Recovery Environment. Except the update throws an unexplained error code for many users: 0x80070643. As it turns out the error is caused by MS changing how WinRE is updated from now on: cumulative updates instead of small patches. Except for many users when they originally installed Windows 10, the installer was not prepared for this so the automatically created Recovery partition is not large enough to apply this update out of no fault of their own.

Since then MS has issued a manual workaround for installing the update. But the workaround involves resizing your system and recovery partitions, so it is only recommended if you are desperate to have this update installed and know what you are doing.

Most people were hoping that in the February patch Tuesday MS would sort out the issue properly, but unfortunately that's doesn't seem to be the case. The erroneous patch hasn't been pulled and no updated version was issued. It is unclear when can we expect MS to properly fix the issue instead of a workaround that is risky even for enthusiasts.

Until now the issue was a mere inconvenience, but since you can't opt out of the broken update even if you don't use Bitlocker, it can interfere with the installation of any further updates as well, as it tries to install it first, but fails blocking the entire update process.

Windows 10 KB5034441 is still broken with 0x80070643 error
 
M76 said:
View attachment 635957

During the January update round MS issued a security update for Windows 10 that should fix a security issue which could allow an attacker to bypass Bitlocker encryption using the Windows Recovery Environment. Except the update throws an unexplained error code for many users: 0x80070643. As it turns out the error is caused by MS changing how WinRE is updated from now on: cumulative updates instead of small patches. Except for many users when they originally installed Windows 10, the installer was not prepared for this so the automatically created Recovery partition is not large enough to apply this update out of no fault of their own.

Since then MS has issued a manual workaround for installing the update. But the workaround involves resizing your system and recovery partitions, so it is only recommended if you are desperate to have this update installed and know what you are doing.

Most people were hoping that in the February patch Tuesday MS would sort out the issue properly, but unfortunately that's doesn't seem to be the case. The erroneous patch hasn't been pulled and no updated version was issued. It is unclear when can we expect MS to properly fix the issue instead of a workaround that is risky even for enthusiasts.

Until now the issue was a mere inconvenience, but since you can't opt out of the broken update even if you don't use Bitlocker, it can interfere with the installation of any further updates as well, as it tries to install it first, but fails blocking the entire update process.

Windows 10 KB5034441 is still broken with 0x80070643 error
Click to expand...
My recovery partition is already larger than what they resize it to and it still fails...
 
  • Like
Reactions: ncjoe
like this
Lakados said:
My recovery partition is already larger than what they resize it to and it still fails...
Click to expand...
According to the linked article the larger the partition the more free space is needed. It's weird, IDK how MS managed this:

  • For Windows 10 v2004 or Windows Server 2022: You should have 50 MB of free space if the partition is smaller than 500 MB.
  • For other versions, you should have more than 300 MB of free space when the partition is 500 MB or larger.
  • When the partition is more than 1 GB, it must have at least 1 GB of free space.
Click to expand...
 
Glad I went ahead and increased my partition via powershell less than a week after this BS update was deployed..

(I REFUSE to let errors exist on my pc)
 
well, I have 1.9gig winre partition, and 426MB are being used on drive......... and it
still fails to install....... I just hid the update using wushowhide.........

not going to waste my time on something MS should fix..
 
Windows has been broken for decades, essentially since the scsi bug appeared and they didn't care about fixing it. That was what, 23 years ago? They don't fix things. They just keep marching along, new bells and whistles, to keep the general population thinking that they are 'upgrading' their software, but do anything but. Problem? Post a note on their website, and when they can't fix it, they just stop responding to people who desperately ask for a way to fix a problem. But the problems remain, over 20 years now. It scares me that so much of our military depends on broken operating systems.
 
Upgrading to Windows 11 still seems to be the easiest fix for this.
 
GotNoRice said:
Upgrading to Windows 11 still seems to be the easiest fix for this.
Click to expand...
Is it a fix, tho? Will updating to W11 resize the partition? If not you are just kicking the can down the road.

If MS was a car dealer they'd mess up your oil change and say sorry, we can't fix it, but you can buy the newer model...
 
Am I understanding this correctly that users will either always encounter the error currently due to not having a large enough recovery partition, or not if they have a large enough partition? And that no patching per se resolves this atm?
 
Okatis said:
Am I understanding this correctly that users will either always encounter the error currently due to not having a large enough recovery partition, or not if they have a large enough partition? And that no patching per se resolves this atm?
Click to expand...
That is correct, if you don't have enough free space on the recovery partition (which increases with the size of the partition), then it will keep failing.
 
M76 said:
View attachment 635957

During the January update round MS issued a security update for Windows 10 that should fix a security issue which could allow an attacker to bypass Bitlocker encryption using the Windows Recovery Environment. Except the update throws an unexplained error code for many users: 0x80070643. As it turns out the error is caused by MS changing how WinRE is updated from now on: cumulative updates instead of small patches. Except for many users when they originally installed Windows 10, the installer was not prepared for this so the automatically created Recovery partition is not large enough to apply this update out of no fault of their own.

Since then MS has issued a manual workaround for installing the update. But the workaround involves resizing your system and recovery partitions, so it is only recommended if you are desperate to have this update installed and know what you are doing.

Most people were hoping that in the February patch Tuesday MS would sort out the issue properly, but unfortunately that's doesn't seem to be the case. The erroneous patch hasn't been pulled and no updated version was issued. It is unclear when can we expect MS to properly fix the issue instead of a workaround that is risky even for enthusiasts.

Until now the issue was a mere inconvenience, but since you can't opt out of the broken update even if you don't use Bitlocker, it can interfere with the installation of any further updates as well, as it tries to install it first, but fails blocking the entire update process.

Windows 10 KB5034441 is still broken with 0x80070643 error
Click to expand...
Yep, I came across this problem and their manual fix did work for me. I'd think you could fix it with a bootable third party partition manager.
 
M76 said:
View attachment 635957

During the January update round MS issued a security update for Windows 10 that should fix a security issue which could allow an attacker to bypass Bitlocker encryption using the Windows Recovery Environment. Except the update throws an unexplained error code for many users: 0x80070643. As it turns out the error is caused by MS changing how WinRE is updated from now on: cumulative updates instead of small patches. Except for many users when they originally installed Windows 10, the installer was not prepared for this so the automatically created Recovery partition is not large enough to apply this update out of no fault of their own.

Since then MS has issued a manual workaround for installing the update. But the workaround involves resizing your system and recovery partitions, so it is only recommended if you are desperate to have this update installed and know what you are doing.

Most people were hoping that in the February patch Tuesday MS would sort out the issue properly, but unfortunately that's doesn't seem to be the case. The erroneous patch hasn't been pulled and no updated version was issued. It is unclear when can we expect MS to properly fix the issue instead of a workaround that is risky even for enthusiasts.

Until now the issue was a mere inconvenience, but since you can't opt out of the broken update even if you don't use Bitlocker, it can interfere with the installation of any further updates as well, as it tries to install it first, but fails blocking the entire update process.

Windows 10 KB5034441 is still broken with 0x80070643 error
Click to expand...
This is already been documented here.

https://hardforum.com/threads/kb5034441-wont-install-with-error-code-0x80070643.2032749/
 
Instead of increasing the size of recovery partition, which I could not do, I simply relocated it to C:\Recovery\
 
Windows has suffered a lot of bugs in general, but hey at least they're implementing new features that nobody will care about or use!

Still remember when I was using win10 the clock being bugged perpetually, no reinstall fixed it. Wouldn't work automatically, had to set it manually. It's a bloody clock, come on. And that's just one.
 
THRESHIN said:
Still remember when I was using win10 the clock being bugged perpetually, no reinstall fixed it. Wouldn't work automatically, had to set it manually. It's a bloody clock, come on. And that's just one.
Click to expand...
It wasn't that the RTC battery was dead?
 
Okatis said:
It wasn't that the RTC battery was dead?
Click to expand...
No. The auto sync from internet was totally broken. This is one example. Another was that the monitors refused to go to sleep mode. Apparently windows thought someone was using the mouse when they weren't. Found a work around, but shit this is pretty simple stuff. I shouldn't have to.

Bottom line is MS has been so busy adding useless features that they forgot to fix the bugs.
 
  • Like
Reactions: Wat
like this
M76 said:
Meanwhile at MS: Another broken update - https://www.ghacks.net/2024/02/27/t...urity-update-fails-to-install-for-some-users/

And ofc there is still no official fix for this one either.
Click to expand...

I'd be very curious to know exactly what specific circumstances have to occur to cause that problem. What I do know is that the article is talking about a cumulative update that was released a week ago at this point, and during this last week I've done updates on at least 20 different Windows 11 computers (in addition to lots of VMs). Not a single one of them had this error.

If I did encounter this error, rather than trying to find some convoluted workaround, I'd probably just do a 10-15 minute in-place upgrade to the latest version which would put you in the exact same place as you would be after installing the cumulative update.

THRESHIN said:
Another was that the monitors refused to go to sleep mode. Apparently windows thought someone was using the mouse when they weren't. Found a work around, but shit this is pretty simple stuff. I shouldn't have to.
Click to expand...

99.9% of the time this is a mouse hardware issue, not anything having to do with Microsoft. Either an old ghetto mouse, dirty sensor, poor choice of mousing surface, or some combo of all three that ends up causing the cursor to jiggle or otherwise not remain perfectly still. How you could possibly blame this on Microsoft blows my mind.
 
GotNoRice said:
I'd be very curious to know exactly what specific circumstances have to occur to cause that problem.
Click to expand...
Just because you don't encounter something doesn't mean the problem is not widespread. You can't do an in-place upgrade if you are already on the latest version of windows. Also not sure if that would be a guaranteed fix.
GotNoRice said:
99.9% of the time this is a mouse hardware issue, not anything having to do with Microsoft. Either an old ghetto mouse, dirty sensor, poor choice of mousing surface, or some combo of all three that ends up causing the cursor to jiggle or otherwise not remain perfectly still. How you could possibly blame this on Microsoft blows my mind.
Click to expand...
I have had the same issue with multiple mice, and it's not that the mouse is moving on its own or drifting (I'd cut my wrists if the $200 mouse was drifting). The mouse itself goes to sleep mode, so it is out of the question that's the problem. The only way I can get the monitor to properly time out is if I disable the ability in device manager for the mouse to wake up the computer. I couldn't even manually put the computer to sleep before I did this, as it would just immediately wake up again.
 
GotNoRice said:
99.9% of the time this is a mouse hardware issue, not anything having to do with Microsoft. Either an old ghetto mouse, dirty sensor, poor choice of mousing surface, or some combo of all three that ends up causing the cursor to jiggle or otherwise not remain perfectly still. How you could possibly blame this on Microsoft blows my mind.
Click to expand...
Linus did a video on Windows terrible sleep mode problems. One of the reasons I've switched to Linux was because of issues with Windows sleep. In my case it was using an older Creative Sound card or an older Asus Xonar sound card. The Creative sound card would sometimes not have any sound when I resumed sleep, and would have to resleep the machine to get back sound. In the Asus Xonar case, the PC would sometimes crash from wake. The community behind UniXonar drivers did have solutions, but they didn't work for me. On Linux these issues are completely gone.

View: https://youtu.be/OHKKcd3sx2c?si=1kb45VkW4HXAQ2Ay
 
GotNoRice said:
I'd be very curious to know exactly what specific circumstances have to occur to cause that problem. What I do know is that the article is talking about a cumulative update that was released a week ago at this point, and during this last week I've done updates on at least 20 different Windows 11 computers (in addition to lots of VMs). Not a single one of them had this error.

If I did encounter this error, rather than trying to find some convoluted workaround, I'd probably just do a 10-15 minute in-place upgrade to the latest version which would put you in the exact same place as you would be after installing the cumulative update.



99.9% of the time this is a mouse hardware issue, not anything having to do with Microsoft. Either an old ghetto mouse, dirty sensor, poor choice of mousing surface, or some combo of all three that ends up causing the cursor to jiggle or otherwise not remain perfectly still. How you could possibly blame this on Microsoft blows my mind.
Click to expand...

same no issues with that new "issues", and the machines ive updated today, didnt even offer the topic update.
and yeah mouse twitch has been an issues for years...
 
M76 said:
You can't do an in-place upgrade if you are already on the latest version of windows.
Click to expand...

If you have not installed the latest cumulative update yet, then you're not on the latest version of Windows.

If you wanted to do an in-place upgrade from Windows 11 23H2 version 22631.3155 to version 22631.3235 for example, which would more or less replicate what occurs when you install the latest cumulative update, you could absolutely do that. The only thing you can't do is go backward; you can't do an in-place upgrade to a version with an older version number than what you already have installed.

M76 said:
The only way I can get the monitor to properly time out is if I disable the ability in device manager for the mouse to wake up the computer.
Click to expand...

So... you disabled the mouse and the problem went away. Sounds pretty conclusive to me. It might not have been a sensor issue on your mouse but it still sounds like it was clearly related to some part of your mouse hardware or mouse software and not the OS.

DukenukemX said:
In my case it was using an older Creative Sound card or an older Asus Xonar sound card. The Creative sound card would sometimes not have any sound when I resumed sleep, and would have to resleep the machine to get back sound.
Click to expand...

Not sure what issue you are having but I'm running 10-15+ year old X-Fi cards in pretty much every desktop computer that I actively use at this point, both PCI and PCIe variants, and it has not caused issues with sleep. Worst thing that happens is that sometimes my EQ settings get set back to stock after an update.
 
Last edited:
DukenukemX said:
Linus did a video on Windows terrible sleep mode problems. One of the reasons I've switched to Linux was because of issues with Windows sleep. In my case it was using an older Creative Sound card or an older Asus Xonar sound card. The Creative sound card would sometimes not have any sound when I resumed sleep, and would have to resleep the machine to get back sound. In the Asus Xonar case, the PC would sometimes crash from wake. The community behind UniXonar drivers did have solutions, but they didn't work for me. On Linux these issues are completely gone.
Click to expand...
I can't speak to Asus's driver quality, but Creative's drivers for cards prior to the X-Fi series always gave me trouble and had a habit of causing BSOD's in XP. Their Vista/7 drivers couldn't crash the computer, but that was because MS forced them into user space, but I would occasionally lose audio and it was only fixable with a restart. I used Daniel_K's drivers for a while (For my Audigy, then Audigy 2 cards) and there was a utility that could cycle the device channels and fix it without a restart. My X-Fi did fine with Win 7 and 10 until I switched to a Zxr which has been trouble free the whole time I've had it.
 
DukenukemX said:
Linus did a video on Windows terrible sleep mode problems. One of the reasons I've switched to Linux was because of issues with Windows sleep. In my case it was using an older Creative Sound card or an older Asus Xonar sound card. The Creative sound card would sometimes not have any sound when I resumed sleep, and would have to resleep the machine to get back sound. In the Asus Xonar case, the PC would sometimes crash from wake. The community behind UniXonar drivers did have solutions, but they didn't work for me. On Linux these issues are completely gone.

View: https://youtu.be/OHKKcd3sx2c?si=1kb45VkW4HXAQ2Ay
Click to expand...


Ya, when i moved to Linux full time, even on my Asus zephyr laptop, everything works out of the box, apps work that I need, no more (knock on wood) silly little things. Heck, even using Steam with their beta library to natively play windows games, runs pretty dam smooth! So now i dont even have a need for a dual boot windows.
 
GotNoRice said:
I'd be very curious to know exactly what specific circumstances have to occur to cause that problem. What I do know is that the article is talking about a cumulative update that was released a week ago at this point, and during this last week I've done updates on at least 20 different Windows 11 computers (in addition to lots of VMs). Not a single one of them had this error.

If I did encounter this error, rather than trying to find some convoluted workaround, I'd probably just do a 10-15 minute in-place upgrade to the latest version which would put you in the exact same place as you would be after installing the cumulative update.



99.9% of the time this is a mouse hardware issue, not anything having to do with Microsoft. Either an old ghetto mouse, dirty sensor, poor choice of mousing surface, or some combo of all three that ends up causing the cursor to jiggle or otherwise not remain perfectly still. How you could possibly blame this on Microsoft blows my mind.
Click to expand...
This was a known and common bug on windows 10. Go ahead, look it up. I'll wait here.
 
GotNoRice said:
I'd be very curious to know exactly what specific circumstances have to occur to cause that problem. What I do know is that the article is talking about a cumulative update that was released a week ago at this point, and during this last week I've done updates on at least 20 different Windows 11 computers (in addition to lots of VMs). Not a single one of them had this error.

If I did encounter this error, rather than trying to find some convoluted workaround, I'd probably just do a 10-15 minute in-place upgrade to the latest version which would put you in the exact same place as you would be after installing the cumulative update.



99.9% of the time this is a mouse hardware issue, not anything having to do with Microsoft. Either an old ghetto mouse, dirty sensor, poor choice of mousing surface, or some combo of all three that ends up causing the cursor to jiggle or otherwise not remain perfectly still. How you could possibly blame this on Microsoft blows my mind.
Click to expand...
I’ve seen this on a collection on our machines, the devices with good SSD’s or NVME’s have no issue, all the ones with cheap or old SSDs fail on this and a variety of others. Anecdotal at best, but that’s all I have been able to figure out, I haven’t bothered with trying to fix it, just upgrade the drives in those units and rebuild suddenly the problems are solved.
 
GotNoRice said:
Not sure what issue you are having but I'm running 10-15+ year old X-Fi cards in pretty much every desktop computer that I actively use at this point, both PCI and PCIe variants, and it has not caused issues with sleep. Worst thing that happens is that sometimes my EQ settings get set back to stock after an update.
Click to expand...
I've had terrible luck with Creative sound cards. They sound the best but they do have problems. I also had an X-Fi SB0880, but the front headphone port just stopped working. My pride and joy was the Recon3D, which did have the resume sleep and losing sound issue. The Recon3D issues might have been related to me using a Ryzen CPU, since I heard there might be issues with the sound card. That too ended up not having working front headphone port, and I even found the shorted component but nobody at badcaps can figure out what the component is. The funny thing was before I switched to Creative, I had the Asus Xonar and just ended up going back to it. The SB0880 didn't have any issues in Windows 10, from what I can remember. A lot of the issues with Windows 10 was the lack of driver updates from the manufacturers. Clearly a driver update could fix the issues and the Recon3D was just a cheaper Sound Blaster Z SE, which did have newer drivers for Windows 10. But this is pretty much the issues with most older sound cards, with the exception of Realtek which also significantly slowed down their driver updates. The Asus Xonar was perfectly fine with Windows 10, until I did an update and then crashing on resume issues. It was a nightmare having to wake up my PC because I was afraid I'd end up rebooting the system anyway, and losing everything I was doing.
 
I am having this issue with Windows Sever 2022 and Azure. There is no WinRe partition in Azure and this updates fails without it. What a stupid thing for Microsoft to do. Its KB5034439 for servers.
 
relapse808 said:
I am having this issue with Windows Sever 2022 and Azure. There is no WinRe partition in Azure and this updates fails without it. What a stupid thing for Microsoft to do. Its KB5034439 for servers.
Click to expand...
just ignore it, theyll add it to a patch later.
 
Last edited:
pendragon1 said:
yeah i did two builds today, it wasnt offered.
Click to expand...
I'm in the process of rolling it out on a random HP that was brought into my office because "its having troubles", troubles with what? I don't even recognize the laptop, I'm pretty sure I am fixing some random person's laptop that was delivered to the admin desk up front... What ever, it has a post-it note with passwords and I'm using it on our isolated network we use for machines we think are infected so this will be fun all the same.
 
DukenukemX said:
I've had terrible luck with Creative sound cards. They sound the best but they do have problems. I also had an X-Fi SB0880, but the front headphone port just stopped working. My pride and joy was the Recon3D, which did have the resume sleep and losing sound issue. The Recon3D issues might have been related to me using a Ryzen CPU, since I heard there might be issues with the sound card. That too ended up not having working front headphone port, and I even found the shorted component but nobody at badcaps can figure out what the component is. The funny thing was before I switched to Creative, I had the Asus Xonar and just ended up going back to it. The SB0880 didn't have any issues in Windows 10, from what I can remember. A lot of the issues with Windows 10 was the lack of driver updates from the manufacturers. Clearly a driver update could fix the issues and the Recon3D was just a cheaper Sound Blaster Z SE, which did have newer drivers for Windows 10. But this is pretty much the issues with most older sound cards, with the exception of Realtek which also significantly slowed down their driver updates. The Asus Xonar was perfectly fine with Windows 10, until I did an update and then crashing on resume issues. It was a nightmare having to wake up my PC because I was afraid I'd end up rebooting the system anyway, and losing everything I was doing.
Click to expand...

Nothing new there. Creative XP drivers were so bad that MS moved the audio system out of the kernel in Vista to eliminate one of the biggest BSOD sources their telemetry was seeing. It was subsequently moved back in for power consumption (less user-kernel switch overhead I assume) as laptop runtime became more important and (I suspect) DRM related reasons with additional hardening around the driver API boundaries to make it less likely for them to be able to trigger a BSOD. (It probably helped here that creative's share shrank a lot over the time period, so any remaining problems would impact fewer people.)
 
GoodBoy said:
For those with failures, are you UEFI boot, or Legacy bios?
Click to expand...
Interesting thought. I haven't found anywhere that says it matters. My one win10 PC is likely a legacy BIOS but I'm not sure. I haven't tried to fix it either. I did run across one posting that claims enabling TPM fixed their machine.
 
Lakados said:
Looks like it has been replaced and fixed with KB5034843
Click to expand...
Nope, the broken update is still broken. That one shows up as an optional quality update, not as a security update.
 
You must log in or register to reply here.
Back
Top