hi, i hope you can help me, recently i install mac os monterrey and my VMware start to show this fault: "Transport (VMDB) error-14: Pipe connection has been broken" I already install the software again and create a new virtual machine, but this message is still shown when I run my virtual machine.
I attach the .log, please, can you help me! I really appreciate it!!
This is a compatibility issue between Fusion 12.2 and macOS 12.0.1, and only on machines with integrated and discrete graphics.
When Fusion starts a VM it requests the discrete GPU, but for some reason Monterey doesn’t grant it. The workaround is to go into system preferences, battery, and disable “Automatic Graphics Switching”
You can re-enable it after starting the VM. I’m sure the next update to Fusion will fix this bug.
I'm having same issue, VMWare 12.2 and Monterey 12.0.1.
In my case I haven't macbook with battery, I have Mac Mini and I haven't Battery option, and I haven't Automatic Switching Graphics, Mac Mini doesn't have it options.
Please, any trick to run virtual machine?, disabling status battery transfer from VM options it doesn't work either.
Thanks.
Jose.
I'm having same issue using MacBook Pro 2012, VMWare 12.2.0 and MacOS Monterey 12.0.1.
When will the fix for this issue be released, has anyone gotten it?
I've had the same issue since upgrading my 2013 Mac Pro to Monterey. The issue happened with Fusion 12.2.0, and still happens now that I've installed 12.2.1. This bug is a blocker for me as I cannot use the disable auto graphics switching workaround—the 2013 Mac Pro has two discrete graphics cards, no integrated card, and no option to disable automatic graphics switching.
I had high hopes that the next fix would address the issue, and since it has not I'm out of ideas and desperate for any workaround or fix. I'm currently using another machine for my Windows tasks, which is incredibly inconvenient. Does anyone have any ideas or suggestions?
Hi,
The opening post has this snippet in the supplied log:
2021-10-27T13:16:51.956Z In(05) mks Metal Device plugin: AppleIntelHD4000GraphicsMTLDriver 16.4.5 10
2021-10-27T13:16:51.956Z In(05) mks Metal Device: Pull Model Interpolation support: No
2021-10-27T13:16:51.956Z In(05) mks MTLRenderer: Enabling workaround for depth zeroing
2021-10-27T13:16:51Z[+0.000] Wa(03) mks Caught signal 6 -- tid 13466 (addr 7FF80DF0F112)
2021-10-27T13:16:51Z[+0.000] In(05) mks SIGNAL: rip 0x7ff80df0f112 rsp 0x70000dc48718 rbp 0x70000dc48740
Which indeed makes it look like the crash is GPU related (check your log if you have this issue as the error itself can mean other things too)
For a GPU related issue like this, chances are you can workaround it by disabling 3D acceleration in your VM configuration.
With the VM shut down go to menu: Virtual Machine -> settings -> Display
and uncheck the Accelerate 3D graphics option down there.
You might be able to tweak the settings there too in some configurations.
Hope this helps,
--
Wil
Unfortunately disabling 3D accel in the VM settings doesn’t fix it. Fusion is still using Metal to scale and render the VM screen, and it does not like certain GPUs at the moment. I’ll check my own logs and see if it only occurs when enabling the depth zeroing workaround:
Integrated GPU (crash):
2021-11-11T17:36:12.537Z In(05) mks MKS-RenderMain: Starting MTLRenderer
2021-11-11T17:36:12.579Z In(05) mks Metal Device name: Intel HD Graphics 4000
2021-11-11T17:36:12.579Z In(05) mks Metal Device removable: FALSE
2021-11-11T17:36:12.579Z In(05) mks Metal Device headless: FALSE
2021-11-11T17:36:12.579Z In(05) mks Metal Device lowPower: TRUE
2021-11-11T17:36:12.579Z In(05) mks Metal Device maxThreadsPerThreadgroup: 1024x1024x1024
2021-11-11T17:36:12.579Z In(05) mks Metal Device recommendedMaxWorkingSetSize: 1610612736
2021-11-11T17:36:12.579Z In(05) mks Metal Device depth24Stencil8PixelFormatSupported: FALSE
2021-11-11T17:36:12.579Z In(05) mks Metal Device GPU family: 1
2021-11-11T17:36:12.579Z In(05) mks Metal Device Apple GPU family: 0
2021-11-11T17:36:12.579Z In(05) mks Metal Shading Language version: 2.4
2021-11-11T17:36:12.579Z In(05) mks Metal RW Texture Tier: 1
2021-11-11T17:36:12.579Z In(05) mks Metal Arg Buffer Support: Tier 1
2021-11-11T17:36:12.580Z In(05) mks Metal Device PCI ID: 8086:0166
2021-11-11T17:36:12.580Z In(05) mks Metal GPU chip: Ivy Bridge (Gen7)
2021-11-11T17:36:13.333Z In(05) mks Metal Device plugin: AppleIntelHD4000GraphicsMTLDriver 16.4.5 10
2021-11-11T17:36:13.333Z In(05) mks Metal Device: Pull Model Interpolation support: No
2021-11-11T17:36:13.333Z In(05) mks MTLRenderer: Enabling workaround for depth zeroing
2021-11-11T17:36:13Z[+0.000] Wa(03) mks Caught signal 6 -- tid 2137492 (addr 7FF8016F6112)
...(registers and stack cut)...
2021-11-11T17:36:13Z[+0.000] In(05) mks SymBacktrace[0] rip=000000010900c59f in function (null) in object /Applications/VMware Fusion.app/Contents/Library/mksSandbox loaded at 0000000108dea000
2021-11-11T17:36:13Z[+0.000] In(05) mks SymBacktrace[1] rip=0000000108fb437f in function (null) in object /Applications/VMware Fusion.app/Contents/Library/mksSandbox loaded at 0000000108dea000
2021-11-11T17:36:13Z[+0.000] In(05) mks SymBacktrace[2] rip=00007ff801741e2d in function _sigtramp in object /usr/lib/system/libsystem_platform.dylib loaded at 00007ff80173e000
2021-11-11T17:36:13Z[+0.000] In(05) mks SymBacktrace[3] rip=00007ff8434f2920 in function OBJC_CLASS_$_MTLArgumentInternal in object /System/Library/Frameworks/Metal.framework/Versions/A/Metal loaded at 00007ff80a49b000
...
Discrete GPU (works):
2021-11-11T17:39:40.657Z In(05) mks MKS-RenderMain: Starting MTLRenderer
2021-11-11T17:39:40.673Z In(05) mks Metal Device name: NVIDIA GeForce GT 650M
2021-11-11T17:39:40.673Z In(05) mks Metal Device removable: FALSE
2021-11-11T17:39:40.673Z In(05) mks Metal Device headless: FALSE
2021-11-11T17:39:40.673Z In(05) mks Metal Device lowPower: FALSE
2021-11-11T17:39:40.673Z In(05) mks Metal Device maxThreadsPerThreadgroup: 1024x1024x64
2021-11-11T17:39:40.673Z In(05) mks Metal Device recommendedMaxWorkingSetSize: 1073741824
2021-11-11T17:39:40.673Z In(05) mks Metal Device depth24Stencil8PixelFormatSupported: TRUE
2021-11-11T17:39:40.673Z In(05) mks Metal Device GPU family: 1
2021-11-11T17:39:40.673Z In(05) mks Metal Device Apple GPU family: 0
2021-11-11T17:39:40.673Z In(05) mks Metal Shading Language version: 2.4
2021-11-11T17:39:40.673Z In(05) mks Metal RW Texture Tier: 1
2021-11-11T17:39:40.673Z In(05) mks Metal Arg Buffer Support: Tier 2
2021-11-11T17:39:40.674Z In(05) mks Metal Device PCI ID: 10DE:0FD5
2021-11-11T17:39:40.674Z In(05) mks Metal GPU chip: Kepler (Kepler)
2021-11-11T17:39:40.805Z In(05) mks Metal Device plugin: GeForceMTLDriver 16.0.12 355.11.11.10.10.143
2021-11-11T17:39:40.805Z In(05) mks Metal Device: Pull Model Interpolation support: No
2021-11-11T17:39:40.926Z In(05) mks Started Shim3D
2021-11-11T17:39:40.926Z In(05) mks MKS-RenderMain: Started MTLRenderer
so yeah seems like it has something to do with that workaround.
Nothing yet? I have the same problem, if you fix it, please ....
Here I haven't made any progress yet, has anyone advanced?
I have been and continue to be completely unable to use Fusion due to this issue (and not having the time to devote to downgrading off of Monterey). I've tried everything I've seen as suggestions, and a few ideas of my own, all to no avail.
It's really laughable at this point: I'm 100% unable to use Windows apps that I've been using for ages, for weeks now, with no clear indication that the issue is even being worked as a matter of urgency.
It is also happening on Intel / Big Sur.
I had one VM opened for 2 weeks since updating from Catalina, had to also upgrade from Fusion 11 to 12.
There was one compatibility issue that I had to fix to make Fusion 12 work.
Quit Fusion after 2 weeks.
On re-opening that VM, the -14 error has come up.
Checked all my other VMs - same error.
So it's not a just Monterey issue and it is happening on Intel (there was a reference to whether it was an M1 issue).
I was able to shutdown a VM (hold down option key in the Virtual Machine menu) and then had access to Settings. Tried toggling the 3D acceleration on and off but it made no difference. I am seeing a bunch of errors in the log that are trying to find the en folder for the messages - inspecting the package the english folder doesn't exist, although the other locales are present. But the logs continue claiming eventually that it couldn't find a "monitor" and needs to abort. (I'll post a log shortly).
Given the number of people posting to this, it is a pretty serious / critical issue - we are all suffering total down time.
Also having the same issue on a 16" 2021 Macbook Pro running Monterey 12.0.1 and Fusion 12.2.1 when importing an existing VM or trying to create a new one. I have tried turning off 3D acceleration, it doesn't make a difference. Still using my old 2019 Macbook Air running Big Sur until this is resolved. This is a less than optimal blocker that seems to be impacting a small number of users unfortunately.
This is what worked for me
sudo kextutil /Applications/VMware\ Fusion.app/Contents/Library/kexts/vmmon.kext
It will trigger some authentication modals.
Hi,
@MitchInSeattle wrote:
Also having the same issue on a 16" 2021 Macbook Pro running Monterey 12.0.1 and Fusion 12.2.1 when importing an existing VM or trying to create a new one.
That's an M1 machine isn't it?
If so then you can't run Fusion 12.2.1 on it as you'll need to run the Apple Silicon Tech Preview instead.
This can be found here: https://communities.vmware.com/t5/Fusion-for-Apple-Silicon-Tech/bd-p/160004
Note however that you can't bring ANY old intel based VM's to your fancy new M1 machine.
No intel virtual machines will run now (or in the future) on your new apple hardware.
That would require emulation of the processor and that's not something that VMware (or any of its competitors) offer. Rosetta also does not help you here as it is too low level for it to take care of that.
You will have to create new Arm based VM's on the M1 machine and the only type of OS that is currently supported on the Tech Preview is Linux.
--
Wil
I tried the sudo kextutil command suggested by duxed; I authenticated in the Terminal for the sudo command, and again in System Preferences to run the new extension, then restarted... and the issue remains. As a reminder I'm running a 2013 Mac Pro (cylinder, aka "trashcan") which worked fine under the latest Big Sur version but stopped working the minute I upgraded to Monterey and hasn't worked since.
ok - so it is NOT the M1 chip - which worried the heck out of me - it IS software based. Good to know but we still have NO fix. How can we accelerate this issue.
You'll get a similar error if you try to run 12.x on an M1 chip. You have to use the tech preview for that.
@ColoradoMarmot wrote:
You'll get a similar error if you try to run 12.x on an M1 chip. You have to use the tech preview for that.
Same error even.
That's the problem with this particular error, it's utterly vague and can mean a lot of things.
What it is basically trying to say is "your VM crashed while trying to start, something went wrong and we can't connect to it"
What the "something" is might be different for everyone in this thread.. which makes it hard to help troubleshooting. Most useful are the mksSandbox.log and the vmware.log, both of which can be found in the VM's bundle.
If those 2 files then gives other users like us a proper hint.. isn't certain, but without that it will be guess work for sure.
It should hopefully give a vmware employee enough to pinpoint what is happening.
--
Wil