PSA another broken Microsoft Patch: KB5087424 (May 2026 hotpatch) breaks 32-bit printing on Server 2022 — splwow64.exe 0xc0000142
Posted by titain19@reddit | sysadmin | View on Reddit | 22 comments
This needs more visibility. Microsoft just wasted 6 hours of my life with an untested patch.
I run an Azure Server 2022 RDS host serving a business application. It suddenly started throwing:
splwow64.exe - Application Error: The application was unable to start correctly (0xc0000142)
Any 32-bit app that touched printing would crash on launch (reproduced it with plain 32-bit Notepad too). 64-bit printing worked fine.
Wasted hours chasing the print driver, VC++ redists, SFC/DISM (all clean) — none of it was the cause.
Root cause: Process Monitor (filter splwow64.exe) showed the process die with exit status -1073741502 (0xc0000142) immediately after touching:
C:\Windows\WinSxS\amd64_microsoft-windows-hotpatches_..._10.0.20348.5074_..._splwow64_hotpatch.exe
Build 20348.5074 = KB5087424, the May 12 2026 Azure hotpatch. The hotpatched splwow64 image fails to map.
Fix: Uninstall KB5087424, reboot. Printing immediately restored. To stop it reinstalling the patch with:
Hide-WindowsUpdate -KBArticleID KB5087424 -Confirm:$false
Not in Microsoft's documented Known Issues yet, but I'm not the only one — there's a Dynamics 365 Community thread of Server 2022 users hitting the identical splwow64 0xc0000142 after KB5087424 and being forced to roll back too.
https://community.dynamics.com/forums/thread/details/?threadid=51c7c262-de52-f111-bec6-7c1e520d540b
jpanixix@reddit
the worst part is Process Monitor was the only thing that actually pointed at the hotpatch.the worst part is Process Monitor was the only thing that actually pointed at the hotpatch. SFC/DISM comes back clean and you're left staring at VC++ redists like an idiot for hours.
kahran@reddit
the worst part is Process Monitor was the only thing that actually pointed at the hotpatch.
justyouropionionman@reddit
The worst part is the hypocrisy...
rswwalker@reddit
All together now in two part harmony!
The worst part is… Process Monitor was… the only thing… that actually pointed at… the hot patch!
raptorshadow@reddit
They say of the Acropolis where the Parthenon is…
Slicker-Firebrand09@reddit
Honestly the worst part is Process Monitor was the only thing that actually pointed at the hotpatch.
Logical-Boat-5161@reddit
I spent all day yesterday trying to figure this out. Thank you for this update.
Doso777@reddit
The question has to be asked: Why 32bit printing?
ScannerBrightly@reddit
Yes. Because old printers didn't stop existing once people moved to 64 bit machines.
RetPala@reddit
Fucking good. Now finish it off.
wisym@reddit
Also breaks generating a PDF
wisym@reddit
My savior! I've been running down a stupid problem for days and this was exactly what I needed.
CharacterLimitHasBee@reddit
Microsoft's way of punishing you for still running 32-bit apps. Or for printing from a server.
Misocainea@reddit
Print servers are a thing though
patchdayalert@reddit
We have no idea what their business constraints may be…could be for a production legacy app?
SarcasticFluency@reddit
It wasted an afternoon for me too. In fact, I was planning on writing up a very similar post this morning once I updated our tech team about the tomfuckery of this BS update. I suspect it was installed for the RDP multi-display scaling issue as it was only installed on the app server by the customer over the weekend, but tough. They want print, they no get RDP fix.
I found it had applied with get-hotfix, but then dug into the files changed and found the same reference to the splwow64.exe hotpatch.
Rubenel@reddit
Systemic brother.....Microsoft fired their Quality Assurance team years ago and deoends on people like us to submit these reports.
Thanks for the report.
patchdayalert@reddit
Appreciate the write-up. I dug into this a bit and wrote up the broader issue here: https://patchdayalert.com/blog/hotpatch-was-supposed-to-be-the-smoother-path/
The frustrating part isn’t just that KB5087424 appears to break splwow64.exe / 32-bit printing, it’s that hotpatch doesn’t seem to give admins a better failure path when something does regress. No reboot sounds great until the rollback is still “uninstall the security update and accept the gap.” For anyone running Azure Edition RDS hosts with legacy 32-bit LOB apps, I’d be testing 32-bit print paths before assuming hotpatch is the safer option. Maybe a more convenient option, but not safer.
HabitAltruistic5648@reddit
Yep. I went through your exact situation a week ago. It was 2 days after launching the new RDS environment in Azure. I was so pissed I rebuilt the machine as non-hotpatch. Never again.
rswwalker@reddit
Instead of hot patch maybe they should call it steaming patch?
New-Alfalfa-2989@reddit
Steaming pile of patch
1stUserEver@reddit
AI said aye aye. Send it.