Microsoft releases Windows 10 patch that fixes audio issues, also breaks printers

midian182

Posts: 9,741   +121
Staff member
Facepalm: Can Microsoft release a bug-fixing update for Windows 10 without it causing new problems? Apparently not, it seems, as yet again the company has released a cumulative update that breaks something else. This time, it’s printers that have been affected.

As reported by Windows Latest, September’s second non-security patch, KB4517211, fixed a number of problems, including one in which audio in certain games was quieter.

Not for the first time, users have reported that the patch has fixed one issue while causing others. Specifically, some HP printers are being affected, though it seems to have hit some other brands, too.

On Microsoft’s community forum and feedback hub, one user wrote: “After installing kb4517211 the print spooler kept crashing. Uninstalling the update and rebooting resolved the issue. Reinstalled to test and print spooler crashes resumed. In addition, the Start button would not work and reported a critical error. Uninstalling the update again fixed the problem. This is on Win 10 1903 RTM.”

Another user on Microsoft’s Answers.com site said they were experiencing similar problems with their HP Officejet Pro 8210 printer after installing the update.

As noted by Techradar, the patch isn’t just interfering with HP printers, though they seem to be most susceptible. Some users have reported issues with PDF printer drivers and Kyocera network printers.

This isn't a new situation for Microsoft, of course. Earlier this month, the company released the KB4515384 patch that addressed a problem with Cortana and spikes in CPU usage. While it did correct the issue, the update resulted in some Start menu searches bringing up blank windows, while others saw a Critical Error warning that informed them 'Your Start menu isn't working. We'll try to fix it the next time you sign in.'

It’s likely that Microsoft’s next patch will fix the printer problem—and probably break something else.

Image credit: Arsenii Palivoda via Shutterstock

Permalink to story.

 
Is there a way to force Windows 7 on Z390? Like a repository of modded ethernet and USB 3.0 drivers?
I think I would move my work and online banking to secondary PC, and risk running my gaming PC on obsolete system that is actually stable.
 
Yep, I got hit with this one myself despite all of the times I said that I've never had any issues. Oh, and the Action Center was also broke with this update. I've since paused updates until sometime in November to prevent this update from coming back until Microsoft fixes their crap.
 
LOL! I read something this morning about some amazing new military vehicle...I think a submarine that uses all new tech with 2 million lines of code (as if that's a good thing). What could go wrong?
 
I think I would move my work and online banking to secondary PC, and risk running my gaming PC on obsolete system that is actually stable.
I think your work and online banking would play nice on Linux. You could go with something easy and familiar like Linux Mint, or a small and fast security focused distro like Alpine Linux.

Other considerations: make a minimal windows 10 installation by combining the tools Windows-ISO-Downloader to get any version of Windows 10 directly from official Microsoft servers (Older versions of Windows 10 are easier to dissect and make minimal), MSMG-Toolkit to remove apps from the ISO, WinToolKit to remove large packages from the ISO, Virtual box to test the ISO, Rufus to flash the ISO to a USB stick. It's actually pretty easy but time consuming and slow. I recommend watching the first YouTube result for "Minimal Windows Install" for some info on what breaks what when removed. Also you can read up on various Windows minimalism themed projects and copy them. These include Windows 10 Decrapifier, Super Minimal Install, Windows 10 Debloater. Any non-removable part of Windows 10 is probably safe to disable with group policy or the registry. Lastly a gem I found in a YouTube comment. Set the group policy "Computer Configuration > Administrative Templates > Windows Components > Windows Update > Configure Updates" to option 2 to never receive an update without you personally choosing to download that specific update.

Another option is to try to track down a legal copy of Window 10 Enterprise LTSB/LTSC. They have no bloat, and receive no feature updates, while getting bug fixes and security updates for 10 years. Keys for activation can be found on eBay, but good luck finding the ISO from a trusted/legal source.
 
LOL! I read something this morning about some amazing new military vehicle...I think a submarine that uses all new tech with 2 million lines of code (as if that's a good thing). What could go wrong?

Keep in mind that integrated systems, especially defense ones, have very rigorous testing regimens that can literally take years to complete end to end. That's part of the reason we spend so much on defense; the overwhelming majority of costs involved are documentation and test, not engineering.

And two million SLOCs is not that much for a modern software project; I've worked on significantly larger systems.
 
Using the update and have a HP 8620, no issue here. Wonder if it's just affecting older models. Hopefully more info will come out of which printers are affected and why.
 
Is there a way to force Windows 7 on Z390? Like a repository of modded ethernet and USB 3.0 drivers?.
As of Intel 1xx chipsets, you theoretically couldn't install Windows 7 directly.

However, Gigabyte's UEFI BIOS has a provision for the BIOS to drive the USB 3.0 until after the OS was installed. The feature was called, "USB 3.0 handoff". It would behoove you to find out if that feature has survived into the latest chipsets.

Again theoretically, you could install Win 7 on boards with PS-2 legacy ports directly,. However most "new" boards only have one PS-2 socket, which is color coded for either/or mouse/keyboard, and you need 2 ports to pull it off.

The lack of USB driver problem did present itself with older style standard BIOS

I thought the ethernet driver was included in the board CD chipset driver and "value added software" package. Did M$ fu*k that up too?.

PS: Please note that these 'solutions', apply only to installing Windows 7 from a DVD. USB sticks create the driver problems.
 
Last edited:
I liked this one better

tired-businesswoman-sleeping-on-table-600w-1321678415.jpg


https://image.shutterstock.com/image-photo/tired-businesswoman-sleeping-on-table-600w-1321678415.jpg
 
Based on the feed back seems that some of the viewers have had a slight more than a tad of a problem with M$......
 
Keep in mind that integrated systems, especially defense ones, have very rigorous testing regimens that can literally take years to complete end to end. That's part of the reason we spend so much on defense; the overwhelming majority of costs involved are documentation and test, not engineering.

And two million SLOCs is not that much for a modern software project; I've worked on significantly larger systems.
Unfortunately, even extensive testing fails - e.g., the recent TS article reporting the woes of touch-screen steering interfaces on US Navy ships.
 
Unfortunately, even extensive testing fails - e.g., the recent TS article reporting the woes of touch-screen steering interfaces on US Navy ships.

Something like that is more an interface/requirement fail; the Navy failed to consider the ability for multiple stations to each independently control ship maneuvering functions, which directly lead to the incident. And rather then address the situation (say, by locking control of these functions to a single station), they are instead going to re-design the ships, which certainly will not raise any other unforeseen problems or costs.

As an example of the difference between military and civilian testing, something akin to the 737 MAX would *never* have seen the light of day if it were given the same scrutiny as military projects.
 
Something like that is more an interface/requirement fail; the Navy failed to consider the ability for multiple stations to each independently control ship maneuvering functions, which directly lead to the incident. And rather then address the situation (say, by locking control of these functions to a single station), they are instead going to re-design the ships, which certainly will not raise any other unforeseen problems or costs.

As an example of the difference between military and civilian testing, something akin to the 737 MAX would *never* have seen the light of day if it were given the same scrutiny as military projects.
I agree that the impression is that the military is rigid and, as such, should not or does not make mistakes. However, no one and no entity is perfect. The military makes mistakes just like everyone else. In the case of the ship steering, it seems that someone should have walked through the workflow in the design phase before it was ever implemented. If they had done so, it seems that they would have realized that such an obvious bug with the system existed.

As to the 737 MAX, I would argue that that would have never happened if the Boeing management was scrupulous instead of being greedy.

I previously worked in the healthcare industry - my day job is as a software engineer. They have rigid testing standards, too, however, that did not prevent software from going out the door with bugs. Sometimes, those rigid standards stand in the way of getting the job done as retesting/recertification, after changes, was (and probably still is) required. I am not saying that testing subsequent releases should not be done, its just that in some circumstances, it is a gargantuan effort, it costs a great deal of time and money to retest and recertify - especially the recertification effort.

I cannot say that the health care software that I worked on always went out the door without bugs; however, as I see it, it is impossible to eliminate bugs from that many lines of code. On the surface, it may appear after testing that there are no bugs, but once users get it and start using it, it's my experience they tend to find things that developers never imagine.

I am also familiar with things like CMM, ISO 9000, and the like. One company I worked for noticed the fact that it was OK to have bugs in the software - if they were documented. From my experience, there are holes everywhere from specification verification, design verification, to testing and validation. It's not an easy task to plug them all. Software going out the door with no known bugs does not imply software going out the door without any bugs.
 
Thank you so much for this article! Our HP printers stopped working...after spending all day trying to fix it, we thought it may be a Windows Update issue...now we know it was! Thanks again!
 
Back