ISBRendererComm : Lost connection to mksSandbox and MKS -RendererMain : Failed to start the renderer

[StefanoK]

Hi all,

I just started using a new virtual machine on my VMware Workstation. Its worth to mention that i have other VM on the station and they are working fine but for some reason on this one i got those two errors and some warnings. I tryed to google it but i didnt find any specific answer. I hope you can somehow help me. I posted down below in the attachment sector photos of the errors with logs.

Thank you in advance

Attachment(s)

Photo and logs.zip   10.64 MB 1 version

[StefanoK]

I solved this problem by downgrading to 15.5.6 . Why does that happen on 16.xx.xx no idea

[bluefirestorm]

There was a major change in 16.x in how virtual graphics were handled (sandboxed) and it also added support for DX11 in Windows VMs. The mksSandbox process does not exist in versions prior to 16.x

From the vmware.log it looks the compatibility is set to 12.

2021-03-30T08:41:44.849+02:00| vmx| I005: DICT         virtualHW.version = "12"

Have you tried setting the virtual HW version to 16.x compatibility when running with Workstation Pro 16.x?

 

[ewolfson]

This may be a dead thread, but I had a similar problem and found that simply disabling "Accelerated 3D Graphics" in ((Virtual Machine Settings> Hardware> Display> 3D Graphics)) allowed me to avoid these crashes and specific errors. 

 

[VaneSolutech]

Thank you ! it worked for me.

[Marko6823]

Gracias.. Se resolvio el problema.

 

Saludos

[smalburg]

what a king you are sir

[josifh]

Thanks hsaqallah,

---

I encountered the problem while trying to edit videos with Lightwork. I haven't tried the proposed solution of disabling 3D acceleration but my guess is that it'll work. Although it hurts to disable it!. I hope VMWare addresses it soon.

 

---

worked for me too on VM Player 17. Thanks for sharing.

[r01100011]

Thank you very much for this. I am using 17.0.0 build-20800274, and I got this error. Disabling Accelerate 3D graphics resolved this issue.

I got this problem when I trying to install Windows 11 VM on a Windows 10 host.

[hfcVM]

I ran into this exact problem with VMWare 17.0.1 on Linux Pop!OS 22.  I really thought I was going to have to find another VM solution.  Thanks for this suggestion!!

-HF

[Alexh28]

Funcionó para mi gracias

[RobRatcliff]

I created a clone of a migrated VM to 16.2, but the shrink failed with the same command. As suggested on this thread, I had to disable the 3D graphics to run "sudo vmware-toolbox-cmd disk shrink /".  

[TeamRED]

November update is triggering this exact issue for me in Windows 11.  I can confirm that disabling 3D Acceleration in the VMware settings for my Win11 vm works.

Obviously not an ideal solution because I need acceleration...

[peter_l1]

I also have the same problem, just upgraded to 16.2.1 and still have it. It's when I do a disk shrink then I get:

mks ISBRendererComm: ISBRendererComm Remote Disconnect: socket closed.
mks The vmx lost its connection to the mksSandbox.
mks The mksSandbox process may have crashed or hung.
mks Check the mksSandbox.log and vmware.log for more details.
mks PANIC: ISBRendererComm: Lost connection to mksSandbox (2878)
mks CoreDump_CoreDump: faking exception to get context

 I can consistently reproduce it when I have 3D graphics enabled when doing a disk shrink via command line inside a Windows 10 VM.

"C:\Program Files\VMware\VMware Tools\VMwareToolboxCmd.exe" disk shrinkonly

 As soon as I disabled 3D graphics it stopped crashing.

[oncedeepred]

I get the same error on Ubuntu 20.04.3 with Workstation 16.2.1 and 16.1 before that. Needless to say that it worked like a charm on 15.5.7. For some reason it takes ages to recover memory file from a mounted iSCSI share on 10Gbps network. I can see network I/O spiking back and forth in bursts up to maximum supported bandwidth. By the time the memory dump is restored, MKS sandbox probably timeouts. After the crash, if I click restore from saved snapshot again, it will restore in a few seconds and will work all day without any issues.

2021-11-15T08:09:45.451Z In(05) vmx MainMem: Opened paging file, '/mnt/data/user/[REDACTED]/Machines/win10-pro-msdn/564decae-695d-d993-eee0-2a11d4a6b192.vmem'.
2021-11-15T08:09:45.454Z In(05) vmx MemSched: caller 0 numvm 1 locked pages: num 25338 max 28814336
2021-11-15T08:09:45.454Z In(05) vmx MemSched: locked Page Limit: host 32840648 config 28822528
2021-11-15T08:09:45.454Z In(05) vmx MemSched: minmempct 50 timestamp 1776
2021-11-15T08:09:45.454Z In(05) vmx MemSched: VM 0 min 4293734 max 8488038 shares 8388608 paged 1205460 nonpaged 28246 anonymous 71184 locked 25338 touchedPct 0 dirtiedPct 0 timestamp 1776 vmResponsive is 1
2021-11-15T08:09:45.454Z In(05) vmx MemSched: locked 25338 target 8488038 balloon 0 0 0 swapped 0 0 allocd 0 512 state 0 100
2021-11-15T08:09:45.454Z In(05) vmx MemSched: states: 0 1 : 1 0 : 2 0 : 3 0
2021-11-15T08:09:45.454Z In(05) vmx MemSched: Balloon enabled 1 guestType 0 maxSize 0
2021-11-15T08:09:45.727Z In(05) vmx Progress 0% (none)
2021-11-15T08:09:53.431Z In(05) vmx Progress 1% (none)
2021-11-15T08:09:59.422Z In(05) vmx Progress 2% (none)
2021-11-15T08:10:02.528Z In(05) vmx Progress 3% (none)
2021-11-15T08:10:06.049Z In(05) vmx Progress 4% (none)
2021-11-15T08:10:08.014Z In(05) vmx Progress 5% (none)
2021-11-15T08:10:09.752Z In(05) vmx Progress 6% (none)
2021-11-15T08:10:12.246Z In(05) vmx Progress 7% (none)
2021-11-15T08:10:15.459Z In(05) vmx Progress 8% (none)
2021-11-15T08:10:20.291Z In(05) vmx Progress 9% (none)
2021-11-15T08:10:27.308Z In(05) vmx Progress 10% (none)
2021-11-15T08:10:43.933Z In(05) vmx Progress 11% (none)
2021-11-15T08:11:04.320Z In(05) vmx Progress 12% (none)
2021-11-15T08:11:09.446Z In(05) vmx Progress 13% (none)
2021-11-15T08:11:15.479Z In(05) vmx Progress 14% (none)
2021-11-15T08:11:21.312Z In(05) vmx Progress 15% (none)
2021-11-15T08:11:30.348Z In(05) vmx Progress 16% (none)
2021-11-15T08:11:40.642Z In(05) vmx Progress 17% (none)
2021-11-15T08:11:50.019Z In(05) vmx Progress 18% (none)
2021-11-15T08:11:56.187Z In(05) vmx Progress 19% (none)
2021-11-15T08:12:01.426Z In(05) vmx Progress 20% (none)
2021-11-15T08:12:05.667Z In(05) vmx Progress 21% (none)
2021-11-15T08:12:18.018Z In(05) vmx Progress 22% (none)
2021-11-15T08:12:28.234Z In(05) vmx Progress 23% (none)
2021-11-15T08:12:33.088Z In(05) vmx Progress 24% (none)
2021-11-15T08:12:37.132Z In(05) vmx Progress 25% (none)
2021-11-15T08:12:41.007Z In(05) vmx Progress 26% (none)
2021-11-15T08:12:49.129Z In(05) vmx Progress 27% (none)
2021-11-15T08:13:06.755Z In(05) vmx Progress 28% (none)
2021-11-15T08:13:13.735Z In(05) vmx Progress 29% (none)
2021-11-15T08:13:20.043Z In(05) vmx Progress 30% (none)
2021-11-15T08:13:23.856Z In(05) vmx Progress 31% (none)
2021-11-15T08:13:27.637Z In(05) vmx Progress 32% (none)
2021-11-15T08:13:39.111Z In(05) vmx Progress 33% (none)
2021-11-15T08:13:45.072Z In(05) vmx Progress 34% (none)
2021-11-15T08:13:51.382Z In(05) vmx Progress 35% (none)
2021-11-15T08:13:56.104Z In(05) vmx Progress 36% (none)
2021-11-15T08:14:03.511Z In(05) vmx Progress 37% (none)
2021-11-15T08:14:15.540Z In(05) vmx Progress 38% (none)
2021-11-15T08:14:24.551Z In(05) vmx Progress 39% (none)
2021-11-15T08:14:30.224Z In(05) vmx Progress 40% (none)
2021-11-15T08:14:34.969Z In(05) vmx Progress 41% (none)
2021-11-15T08:14:40.380Z In(05) vmx Progress 42% (none)
2021-11-15T08:14:45.713Z In(05) vmx Progress 43% (none)
2021-11-15T08:14:54.455Z In(05) vmx Progress 44% (none)
2021-11-15T08:15:04.351Z In(05) vmx Progress 45% (none)
2021-11-15T08:15:09.656Z In(05) vmx Progress 46% (none)
2021-11-15T08:15:13.775Z In(05) vmx Progress 47% (none)
2021-11-15T08:15:18.792Z In(05) vmx Progress 48% (none)
2021-11-15T08:15:24.978Z In(05) vmx Progress 49% (none)
2021-11-15T08:15:35.077Z In(05) vmx Progress 50% (none)
2021-11-15T08:15:43.953Z In(05) vmx Progress 51% (none)
2021-11-15T08:15:50.541Z In(05) vmx Progress 52% (none)
2021-11-15T08:15:58.651Z In(05) vmx Progress 53% (none)
2021-11-15T08:16:04.689Z In(05) vmx Progress 54% (none)
2021-11-15T08:16:16.114Z In(05) vmx Progress 55% (none)
2021-11-15T08:16:26.312Z In(05) vmx Progress 56% (none)
2021-11-15T08:16:31.086Z In(05) vmx Progress 57% (none)
2021-11-15T08:16:35.712Z In(05) vmx Progress 58% (none)
2021-11-15T08:16:41.121Z In(05) vmx Progress 59% (none)
2021-11-15T08:16:50.515Z In(05) vmx Progress 60% (none)
2021-11-15T08:17:00.024Z In(05) vmx Progress 61% (none)
2021-11-15T08:17:08.469Z In(05) vmx Progress 62% (none)
2021-11-15T08:17:12.693Z In(05) vmx Progress 63% (none)
2021-11-15T08:17:15.704Z In(05) vmx Progress 64% (none)
2021-11-15T08:17:17.881Z In(05) vmx Progress 65% (none)
2021-11-15T08:17:20.446Z In(05) vmx Progress 66% (none)
2021-11-15T08:17:23.999Z In(05) vmx Progress 67% (none)
2021-11-15T08:17:36.623Z In(05) vmx Progress 68% (none)
2021-11-15T08:17:41.629Z In(05) vmx Progress 69% (none)
2021-11-15T08:17:45.905Z In(05) vmx Progress 70% (none)
2021-11-15T08:17:50.368Z In(05) vmx Progress 71% (none)
2021-11-15T08:17:54.434Z In(05) vmx Progress 72% (none)
2021-11-15T08:17:58.444Z In(05) vmx Progress 73% (none)
2021-11-15T08:18:02.433Z In(05) vmx Progress 74% (none)
2021-11-15T08:18:14.059Z In(05) vmx Progress 75% (none)
2021-11-15T08:18:20.656Z In(05) vmx Progress 76% (none)
2021-11-15T08:18:24.939Z In(05) vmx Progress 77% (none)
2021-11-15T08:18:29.502Z In(05) vmx Progress 78% (none)
2021-11-15T08:18:32.124Z In(05) vmx Progress 79% (none)
2021-11-15T08:18:34.684Z In(05) vmx Progress 80% (none)
2021-11-15T08:18:37.406Z In(05) vmx Progress 81% (none)
2021-11-15T08:18:40.368Z In(05) vmx Progress 82% (none)
2021-11-15T08:18:42.474Z In(05) vmx Progress 83% (none)
2021-11-15T08:18:45.158Z In(05) vmx Progress 84% (none)
2021-11-15T08:19:05.572Z In(05) vmx Progress 85% (none)
2021-11-15T08:19:10.941Z In(05) vmx Progress 86% (none)
2021-11-15T08:19:16.405Z In(05) vmx Progress 87% (none)
2021-11-15T08:19:23.402Z In(05) vmx Progress 88% (none)
2021-11-15T08:19:26.370Z In(05) vmx Progress 89% (none)
2021-11-15T08:19:27.645Z In(05) vmx MainMem: Prefetch 29405 MB from checkpoint (7527904 hot, 5776130 locked, lazy = 1).
....
2021-11-15T08:19:27.877Z In(05) mks SOCKET 4 (100) recv detected client closed connection
2021-11-15T08:19:27.877Z In(05) mks ISBRendererComm: ISBRendererComm Remote Disconnect: socket closed.
2021-11-15T08:19:27.877Z Wa(03) mks The vmx lost its connection to the mksSandbox.
2021-11-15T08:19:27.877Z Wa(03) mks The mksSandbox process may have crashed or hung.
2021-11-15T08:19:27.877Z Wa(03) mks Check the mksSandbox.log and vmware.log for more details.
2021-11-15T08:19:27.877Z Cr(01) mks PANIC: ISBRendererComm: Lost connection to mksSandbox (2878)
2021-11-15T08:19:27.877Z In(05) mks Panic: can't get userlevel lock.
2021-11-15T08:19:27.877Z Wa(03) mks
2021-11-15T08:19:27.877Z Wa(03)+ mks The core dump limit is set to ZERO; no core dump should be expected

---

[baneeishaque]

Workaround is working...

[TimothyHuckabay]

I too encountered this issue, immediately after upgrading to 16.2.1!  So, I downgraded Workstation Pro back to 16.1.2 (what I'd previously had), and my VM is again working!  VMware seems to no longer have good regression testing, suggesting that they and/or the company's software engineers / QA folks are of a lower grade than in the past.  VMware needs to get its act together!

[dagbj]

I encountered  this problem as well. My story is that I upgraded to 16.2.1 from 16.2.0, with a VM that had the HW upgraded.

Disabling 3D accelration enables me to access the VM. It looks to me that So this is a bug introduced with 16.2.1.

My setup is:

Ubuntu 20.04.3 LTS, kernel 5.11.0-40-generic
I'm using secure boot and got the modules signed and all well there.

If I understand things correctly one need the 3D accelration in the VM to have it pass the "win 11 upgrade test".

[Lars_Skjærlund]

I was a happy user of VMware Workstation on OpenSUSE 10 years ago.

Since our ways have parted, and everytime I've evaluated it in order to return to VMware, it fails like this.

I'm now running Ubuntu 21.10, and I've now spent almost a whole day trying to get a useful installation.

I'm running it on a Lenovo Carbon X1 gen. 6 with Core i7 CPU, 16 GB RAM, and 500 GB SSD. I've created a fresh Win10 install with 2 CPUs, 4 GB RAM, 60 GB disk fully allocated from the beginning - and yet you have to wait literally for minutes for response when you click something in the GUI.

Neither this time will VMware win me back as a customer.

[RDPetruska]

What version of Workstation are you using?  

There have been lots of reports of issues with 16.2.0 and 16.2.1.  Have you tried 16.1.2?

[Lars_Skjærlund]

I'm running 16.2.1 build-18811642.

Would a downgrade be worth a try?

[Lars_Skjærlund]

Downgrading to 16.1.2 helped a lot - but my 3D app still runs a lot slower than using Wine directly on Linux.

3D acceleration would surely help.

[Mouri_Naruto]

I have met the same issue when a dialog box pops up too long at startup without selecting by me.

Maybe the mksSandbox thinks the connection to vmware-vmx has been timeouted.

Kenji Mouri

[lytzen]

Looks like version 16.2.2 is out. Has anyone tried to see if that works. I am running out of energy to keep trying new versions, which never work.

[RobRatcliff]

16.2.2 still requires that you disable 3d acceleration before doing the shrink operation. 

[lytzen]

Thanks

[montge]

Confirming issue still exists  with 16.2.1 build-18811642

I wish that VMware's help messages were more helpful.

[pedoc]

vmware 17 same probrom

[hsaqallah]

I encountered the problem while trying to edit videos with Lightwork. I haven't tried the proposed solution of disabling 3D acceleration but my guess is that it'll work. Although it hurts to disable it!. I hope VMWare addresses it soon.

[lytzen]

I might depend on which version of VMware you have installed. Version 16.2.1 does not work for me, even with 3D disabled. I have to run 16.1.2. I have not yet tried version 16.2.2 though..

[gregoliver]

mks is crashing from a panic and says it is creating a coredump, but it does not actually create one:

 

 

2022-02-10T20:40:32.460Z Cr(01) mks  PANIC: VERIFY bora/mks/renderers/vulkan/vkrSwapchain.c:415
2022-02-10T20:40:32.460Z Cr(01)+ mks  
2022-02-10T20:40:32.460Z Wa(03) mks  
2022-02-10T20:40:32.460Z Wa(03)+ mks  Attempting a core dump using an unlimited size limit
2022-02-10T20:40:32.673Z Wa(03) mks  Core dumped.
2022-02-10T20:40:32.673Z In(05) mks  Core dump pipes to process /usr/lib/systemd/systemd-coredump %P %u %g %s %t %c %h, core file unreliable
2022-02-10T20:40:32.673Z Wa(03) mks  A core file is available in "/home/greg/Virtuals/Windows10-Pro/mksSandbox/core.9042"

 

 

 

ls -la /home/greg/Virtuals/Windows10-Pro/mksSandbox/core.9042
ls: cannot access '/home/greg/Virtuals/Windows10-Pro/mksSandbox/core.9042': No such file or directory

Systemd was intercepting it:

Process 9476 (vmx-svga) of user 1000 dumped core.
                                                  
                                                  Found module linux-vdso.so.1 with build-id: f01a5289def9bee2abdf28b4671e1f155f5f1cb8
                                                  Found module libpcsclite.so.1 with build-id: 2a59c60b46b91dbc5e4035a9e7e7a9929b8d642f
                                                  Found module libgpg-error.so.0 with build-id: 081975d0a3416374b4883b2f1639fd3c9df42390
                                                  Found module libresolv.so.2 with build-id: 0fbe5191a8f9212dd09bf0f85b08fcec995d543a
                                                  Found module libgcrypt.so.20 with build-id: 711d41580c5a8649a79a8430a985dac3e25b5ba2
                                                  Found module libcap.so.2 with build-id: 0214aa9cc6a8646eb9ec27ab7dda6a2219da6500
                                                  Found module liblz4.so.1 with build-id: fd02c4542a1ce1ad6c503b958447be1c37a4afee
                                                  Found module libzstd.so.1 with build-id: 7984add4b0ab24869c30e19cfba3637ca295dd7b
                                                  Found module liblzma.so.5 with build-id: 7fec53ce7cba9489f130c99cfd3ace82e9dde0ee
                                                  Found module libogg.so.0 with build-id: 714c2b8484f85a852e216c4fc4bb2cf92872f37e
                                                  Found module libopus.so.0 with build-id: 0717c4aa935190f889e12134adcb95f1173896e6
                                                  Found module libvorbisenc.so.2 with build-id: 68b1fdf0007c752a46482443a9b599bb6cf5a64e
                                                  Found module libvorbis.so.0 with build-id: d3aece982a2dfdf9f412b5fc8a4dc88722bde200
                                                  Found module libFLAC.so.8 with build-id: 34324bdfee454c456c4e08456152fe5b86ba331f
                                                  Found module libgsm.so.1 with build-id: f84c8103f3f95ee09f2c2888e21ff13c870ee056
                                                  Found module libasyncns.so.0 with build-id: 648745a2e1c2a0a990336ca37d634b817a6861c5
                                                  Found module libsystemd.so.0 with build-id: b61753e8440ecf39624b3e37f855941c18a69e9b
                                                  Found module libsndfile.so.1 with build-id: cd7eb73b7b9313a890809eb349ff6322a79d9191
                                                  Found module libdbus-1.so.3 with build-id: 58e800f0865bde326c5c9ce510062d5b19902f67
                                                  Found module libpulsecommon-15.0.so with build-id: 610784314edf8434b1e341c44c97a92afab2a37e
                                                  Found module libpulse.so.0 with build-id: dae2a9442ffad84ff9620bb7cb2ea33517dea455
                                                  Found module libaio.so.1 with build-id: 10d0b58965f676c6d82d836326436912cfa81b7e
                                                  Found module libnuma.so.1 with build-id: 24b3826eb785d8e4dfec3a9b8e5fe37117269d57
                                                  Found module libnss_myhostname.so.2 with build-id: 884539c682a7226ebe838279cb382c786e469e12
                                                  Found module libXau.so.6 with build-id: b6a2d4859848c28bf0a4fc8744d04b1935879af6
                                                  Found module libXfixes.so.3 with build-id: e8020da14577d275b75ed0c4cb18b1050fc33595
                                                  Found module libXrender.so.1 with build-id: 45e135fd2ee2393b7eb0c94f964542b8326582e1
                                                  Found module libxcb.so.1 with build-id: 1352e33391b3079347f320218799d6e531bd42f1
                                                  Found module ld-linux-x86-64.so.2 with build-id: 524f5518285687402d801b9c3b1e61d435e6780b
                                                  Found module libc.so.6 with build-id: d7e3e120ff5fafc298666927c6475567348588ef
                                                  Found module libpthread.so.0 with build-id: ef966cfe5a73990033e35779603c0d78401013e6
                                                  Found module libgcc_s.so.1 with build-id: 88564abce789aa42536da1247a57ff6062d61dcb
                                                  Found module libm.so.6 with build-id: 47343f97c29c4e91c841e2bfd344329a264075ad
                                                  Found module libstdc++.so.6 with build-id: 64dc8524cb0109c7f3d3c0ef3d4d6ec9979db75c
                                                  Found module libXtst.so.6 with build-id: 97ddda29c72fa18b2cae1274026730a04f8b0ae2
                                                  Found module libXcursor.so.1 with build-id: fb3d6eb4c582d4761eab821bc0ceb77fc7c2654f
                                                  Found module libXinerama.so.1 with build-id: 72d169302fe5696536345d6da1605baeb6e3bc01
                                                  Found module libXi.so.6 with build-id: e1d4fcbcab03948da50a0e175ac2459c335e7939
                                                  Found module libXext.so.6 with build-id: 08e82aa07cb7910a08d8386fb16de7896ce290f4
                                                  Found module libX11.so.6 with build-id: b63e8f330442522b1d5872d1b2ccf32702767d80
                                                  Found module libdl.so.2 with build-id: 01afa92b2c758f2702ff6bdc0f6cadeac64456f7
                                                  Found module librt.so.1 with build-id: 7942627f54693fdeb840897c0b959b7fd909f5b3
                                                  Found module vmware-vmx with build-id: 64720c96cb791ce35d5320f4c40ec9ef892bcb34
                                                  Stack trace of thread 9476:
                                                  #0  0x00007fd6c1a87ecd syscall (libc.so.6 + 0x10aecd)
                                                  #1  0x000055b0ad29a88a n/a (vmware-vmx + 0x89a88a)
                                                  #2  0x000055b0ad29aba9 n/a (vmware-vmx + 0x89aba9)
                                                  #3  0x000055b0acb3dab3 n/a (vmware-vmx + 0x13dab3)
                                                  #4  0x000055b0acb237b5 n/a (vmware-vmx + 0x1237b5)
                                                  #5  0x000055b0acefeaea n/a (vmware-vmx + 0x4feaea)
                                                  #6  0x000055b0aceff266 n/a (vmware-vmx + 0x4ff266)
                                                  #7  0x000055b0aceff818 n/a (vmware-vmx + 0x4ff818)
                                                  #8  0x000055b0acf03e0f n/a (vmware-vmx + 0x503e0f)
                                                  #9  0x000055b0acd70349 n/a (vmware-vmx + 0x370349)
                                                  #10 0x000055b0acd714c5 n/a (vmware-vmx + 0x3714c5)
                                                  #11 0x000055b0acd71a34 n/a (vmware-vmx + 0x371a34)
                                                  #12 0x000055b0acd71af0 n/a (vmware-vmx + 0x371af0)
                                                  #13 0x000055b0acd71e2a n/a (vmware-vmx + 0x371e2a)
                                                  #14 0x000055b0accee6ee n/a (vmware-vmx + 0x2ee6ee)
                                                  #15 0x000055b0acdb80be n/a (vmware-vmx + 0x3b80be)
                                                  #16 0x000055b0acdb6431 n/a (vmware-vmx + 0x3b6431)
                                                  #17 0x000055b0acd9c465 n/a (vmware-vmx + 0x39c465)
                                                  #18 0x000055b0accc41e2 n/a (vmware-vmx + 0x2c41e2)
                                                  #19 0x000055b0accc4655 n/a (vmware-vmx + 0x2c4655)
                                                  #20 0x000055b0acdb088d n/a (vmware-vmx + 0x3b088d)
                                                  #21 0x000055b0ad0ee515 n/a (vmware-vmx + 0x6ee515)
                                                  #22 0x00007fd6c1a0ab1a start_thread (libc.so.6 + 0x8db1a)
                                                  #23 0x00007fd6c1a8f650 __clone3 (libc.so.6 + 0x112650)

 

[gregoliver]

It is a Wayland.  When switching back to Xorg, all works as expected.

 

 

[Sylvain-French]

Same issue with 16.2.2 and Debian 11. Crash (even with 3D option disabled) and corrupt the VMDK file. Need to reinstall all.

I've waste 1 day with VMWare and I need to progress in my work. I will switch to another solution.

Previously I used v15 and Debian 10 without problem.

Good luck, but v16x seems to be an important regression.

[josifh]

Wrong reply, sorry. But why i can't delete my posts ?!

[Thorsten200]

Hi all,

In our Company we are useing Siemens Field PG 5 for Programming Software like Simatic Manager and Tia Portal.

We have the same Laptops, but at some colleagues the VM crashes and at some clleagues it doesn't crash.

Workstation 16 Pro

VM Software Version 16.2.2 build-19200509

The VM Tools are up to date.

I read a lot about downgrading to 16.1 / or even 15.5

Could this be really the case?

 

In mksSandbox these following logs are posted.

2022-02-24T12:46:39.782Z In(05) svga ReplayFifo: mksReplay format version=18
2022-02-24T12:46:39.782Z In(05) svga MKSRoleReplay: MKSRoleReplay_Run: Starting.
2022-02-24T12:47:17.348Z Wa(03) svga DX11ShimOps: Intel doesn't support zeroing UAVs with other bind flags
2022-02-24T13:19:52.053Z Wa(03) svga The mksSandbox lost its connection to the main mks.
2022-02-24T13:19:52.053Z Wa(03) svga The main mks process may have crashed or hung.
2022-02-24T13:19:52.053Z Wa(03) svga Check the mksSandbox.log and vmware.log for more details.
2022-02-24T13:19:52.053Z Cr(01) svga PANIC: MKSSandboxComm: Lost connection to isbRenderer (1408)
2022-02-24T13:19:52.053Z In(05) svga CoreDump_CoreDump: faking exception to get context
2022-02-24T13:19:52.059Z In(05) svga CoreDump: Minidump file C:\Users\PC-FDE~1\AppData\Local\Temp\vmware-PC-FDE1-SC11\mkssandbox.dmp exists. Rotating ...
2022-02-24T13:19:52.059Z In(05) svga CoreDump: Current max dump count is 5.
2022-02-24T13:19:52.062Z Wa(03) svga CoreDump: Writing minidump to C:\Users\PC-FDE~1\AppData\Local\Temp\vmware-PC-FDE1-SC11\mkssandbox.dmp
2022-02-24T13:19:52.213Z In(05) svga CoreDump: including module base 0x0x7ff611980000 size 0x0x00d40000

[lytzen]

It seems to be a problem with VMware on windows only. I am running VMware Workstation 16.2.3 om my Mac at home, with 3D graphics and all enabled, with no problems at all.

If I run the exact same virtual machine on my windows 10 PC at work, any VMware Workstation installation newer than 16.1.2 will crash - even with 3D graphics disabled, and it only works on 16.1.2 because I have 3D graphics disabled.

I am considering moving to Virtual Box or qemu, to see if I can get better stability than VMware can provide.

[strudels]

tl;dr for anyone looking for a better/different workaround than disabling 3d acceleration: try switching from Nvidia to Intel integrated graphics. Doing this allowed me to continue using 3d acceleration.

I stumbled upon this issue while upgrading my ubuntu 20.04 kernel from 5.13 to 5.15 via the linux-generic-hwe-20.04 package (i.e. the package that provides the ubuntu 22.04 kernel to ubuntu 20.04). After upgrading, I ran into an issue where vmmon was not compiling correctly with the new kernel 5.15 headers. That issue was described here: https://communities.vmware.com/t5/VMware-Workstation-Pro/VMware-16-2-3-not-working-on-Ubuntu-22-04-LTS/m-p/2905535#M175399

Instead of installing the kernel modules built from the vmware-host-modules github repo linked in that issue, I decided to upgrade my vmware workstation version from 16.1.2 to 16.2.4 and see if the error went away. To my relief, vmmon was able to compile, and I was able to start VMs again.

However, after starting a VM, it would hang for a while, and then eventually show this same `MKS -RendererMain` error. Indeed, disabling 3d acceleration would allow for the VM to work as normal.

Given that I started having these issues originally from my kernel 5.13 to 5.15 upgrade, and that I know ubuntu 22.04 uses both kernel 5.15 and wayland by default, I guessed that the issue of 3d acceleration not working was related to this. In ubuntu 22.04 the stack would be kernel5.15+nvidia+wayland, where my stack was kernel5.15+nvidia+x11. I used `prime-select intel` to switch from my nvidia GPU to my intel integrated graphics, and was able to start my VMs again successfully.

I have a Dell Precision 5510 laptop, with a Nvidia Quadro T1000 Mobile discrete graphics card. My VM was running Windows 10. And as previously stated my host machine was Ubuntu 20.04 w/ linux-generic-hwe-20.04 installed.

[strudels]

Looks like my theory about my issue being related to the wayland upgrade was at least partially right. I ended up finding this article: https://wiki.vi-toolkit.com/index.php?title=Vmx_hacks&mobileaction=toggle_view_desktop#Linux_Host_-_switch_back_from_Vulkan_to_X11

I added this to my `~/.vmware/config` file, and was able to continue to use my Windows 10 VM with my nvidia GPU and 3d acceleration enabled:

```
mks.enableX11Presentation=TRUE
mks.enableVulkanPresentation=FALSE
mks.enableVulkanRenderer = "FALSE"
mks.enableGLRenderer = "TRUE"
mks.vk.allowUnsupportedDevices = "TRUE"
mks.forceDiscreteGPU = "TRUE"
```

[SHoldgaard]

FYI - version 16.2.4 still requires to disable 3D Accellerated graphics. 

[malvcr]

16.2.4 build 20089737 has the same behavior when running  Linux 5.15-46 on 5.15.0 ... both with Wayland (Ubuntu on Ubuntu-Studio) ... however

I noticed that the Virtual Machine had 4GB RAM but the Graphics Memory 8GB ...

Then, without disabling the 3D Graphics, I defined 8GB RAM and 2GB of Graphics Memory and it seems that works.  Just that a remote VNC connection freeze, but could be a different issue.

[acinsa]

Thanks strudels. I added these lines to my vmx on my Debian Buster host with Nvidia Quadro RTX4000 and Windows10 VM. This allowed me to launch the VM without hanging and with 3D acceleration on. Great result and the performance is on a par if not slightly better than with Virtualbox.

 

[omf2]

It's happening on Workstation 16.2.4 running on a Windows 10 host with Intel integrated graphics, so it's not likely to be an NVidia issue, at least not on Windows (which makes it unlikely on Linux, I think).

[hsaqallah]

So I encountered this MKS during upgrading my guest Windows 10 to Windows 11 (although 3D HW Acceleration was turned off.) But the error was happening on a specific VmwarePlayer instance (I have the latest 16 version). Other instances were fine. For the faulty instance, I tried defragmenting and compacting the main vmdk hard drive (from the VMWare Hardware -> Hard Disk screen.) Then I did chkdsk /f /r on the host windows machine on the drive that has my vmdk files. Interestingly, chkdsk found some free space related errors and fixed them. After that, my guest upgrade to windows 11 worked. Not sure what step fixed it. 

[Loren_Pechtel]

Count me amongst those for whom this is a solution.  I'm not normally doing anything in the VMs that would use 3D graphics, the loss is unimportant.

I noticed a strong pattern--it would blow up when the system was otherwise heavily loaded or otherwise not responding well.  (When a program wanted to update itself--I think it did a restore point before updating, took out a VM.)

[tfuto]

For all the folks, who can reproduce this error: Are you sure this is not a Windows/graphics driver issue?

- There is a "feature" in Windows, that unresponsive GPUs are reset after a timeout. (TDR - Timeout Detection and Recovery, https://learn.microsoft.com/en-us/windows-hardware/drivers/display/timeout-detection-and-recovery)

- Shared GPU memory means that GPU memory can actually be paged out.

- During paging, GPU call latency is enormous.

- Observing a call not responding within a reasonable time, Windows can decide that GPU is frozen and resets it.

I have seen this many times, also during excessive (normal) process swapping, too. Then you get MKS disconnection, etc.

When you disable 3D acceleration, you are not using the GPU's resources as heavily.

Could you please try installing the latest graphic card driver, and also change the timeout value? Please see: https://answers.microsoft.com/en-us/windows/forum/all/increase-time-out-limit/e979e2ad-e15f-450b-9818-a148cbf01078

I saw this disconnect many times, and haunted me for awhile. It happened with Win10, RTX2070 and RTX3070Ti, when not using Workstation on Hyper-V. When using Workstation with Hyper-V, these errors got very rare. I cannot reproduce this error myself deterministically. My best way to crash VMware is having a large RAM Ubu22 guest VM (32 GB on a 64 GB machine), that starts with lots of I/O on NVMe, and starting Firefox with a page that uses a DRM video decoder (Netflix is the winner, but DisneyPlus works, too). I checked with the Firefox guys, reported bugs, etc. but it seems that although there is a correlation, the error seems to be an OS issue.

Could you also please look for any errors in Event Viewer after these disconnects? Thank you.

[mstepura]

In case if anyone was hoping the mskSandbox bugs would be fixed in VMWare Workstation 17 - the issue is still there, at least for 17.0.

[nrjank]

in fact, I was running a linux vm on win10 host fine in v16, just upgraded to v17, and started getting this error when trying to boot my vm. disabling graphics acceleration seemed to solve the problem. not ideal. might downgrade back to 16.

[kutewenox]

Hey guys,

I'm assuming it's an error, but after a lot of trying I found the solution:

Add "mks.dx12.vendorID = "0x10de"" to .vmx file and the vm also can use with accelerate 3d graphics. (The vondorID describes the (i)gpuID and shown in the device manager.)

Please try.

[ChrisWV]

Dear Friend,

Could you kindly let me know Where can find the file as you mentioned for add -mks.dx12.vendorID = "0x10de", for solve the problem.

my mail adress at:[e-mail]

thank you,

Chris

 

[TomPham1]

It is in the location where you store the disk (vmdk)

if you use the default then it is in

C:\Users\<UserID>\Documents\Virtual Machines\Windows 10 x64

 

 

 

[Networxs]

Hi there,

encountered same issue when upgrading from VM Workstation Player 16 to 17.0.0 and my WinXP VM could not be started anymore due to subjected Error. While running chkdsk on particular .vmdk did not help, eventually disabling 3D graphics acceleration in particular VM's Hardware > Display settings did.

[BadOmen]

I'm on Workstation 17.0.2 on an HP EliteBook 850 G5.  I'm having the same problem in a Windows 7 VM.  Problem is that disabling 3D acceleration also disables the Windows Aero theme.  This really pisses me off.  For a program that I pay for, I would hope that there would be a better fix for it by now.  It's been two years.

I've even tried downgrading VMware Tools version to 10.3.5, and same thing.  Downgraded the hardware compatibility to version 15.x, same thing.  It's one thing if all I need to do is install the VM as HW version 15 off the bat.  But I have a couple of Win7 VM that I made a couple of years ago, and that would mean that I would have to reinstall them outright (I've upgraded their hardware versions).  One is a cloned physical disk, and I really don't want to have to do that again if it starts having the same problem.

Edit: Sorry for being snarky and crappy.  Was uncalled for.  I had just figured out how to get my Vista VMs to work properly, and then started having problems with Windows 7.  just big grimace.

 

Edit: So far, downgrading hardware version to Workstation 14.x (if 15.x doesn't work) and reducing GPU RAM seems to work so far.  But this should still be fixed.

[Jose Fonseca]

@BadOmen, this is a generic error message -- there is not one issue that triggers this error message -- but lots of different and new issues -- often due to D3D12 driver issues.

To understand and fix your particular problem, we'll need vmware.log and mksSandbox.log files, and a description of what was happening in the guest when the error happened.

Per https://support.hp.com/gb-en/document/c05903028 the HP EliteBook 850 G5 has a Intel UHD Graphics 620 and an optional discrete AMD Radeon R59.  Does your system has the optional AMD GPU?

 , FYI.

[BadOmen]

No it does not.  It's definitely a graphics issue, given as I'll open an app like Chess Titans to test on Win7, and that will do it.

 

Downgrading the HW version of the VM and reducing GPU RAM to 1 or 2GB seems to help so far.

[Michal Krol]

BadOmen,

Can you attach a vmware.log and mksSandbox.log?

[BadOmen]

Hello.  So I built a test VM, upgraded to HW 17.X and most recent VMware Tools.  3D acceleration enabled, and 1GB GPU RAM.  4 Cores CPU, 4GB RAM, SATA drive.  I went in to a built-in Windows game like Chess Titans, and it crashed within a minute.  I attached the files in the zip.

Also, Sorry for being snarky and crappy. Was uncalled for. I had just figured out how to get my Vista VMs to work properly with an older version of VMware Tools, and then started having problems with Windows 7.

Attachment(s)

Windows 7 (test).zip   634 KB 1 version

[BadOmen]

So before upgrading that VM to 17.X, I originally built it for HW 15.X.  So I copied a backup, and ran that.  Crashed as well (logs attached).  However, downgrading to HW 14.X, and no problems so far.  Played some games that used graphics, etc.  No problem.  So it looks like Windows 7 with 3d acceleration enabled needs to stay at HW version 14.X so far.

Attachment(s)

Windows 7 (test)-HW 15.X.zip   630 KB 1 version

[Developer5]

FIXED IN VMWARE WORKSTATION 17.5. PLEASE UPGRADE YOUR SOFTWARE.

[nrjank]

Confirming I was able to reenable hardware acceleration without error

[VMware Community Member]

i have/had the same issue with upgraded W10 to W11 machines. but when my machiens are upgraded to V17.5.x HW, the W11 machines keep freezing with 3d graphics disabled.  when degrading back to V17.x they chrash again. but i cannot add the line of code into the VMX file, because the machines need to be encrypted.

[VMware Community Member]

I also still have this issue on VMWarePlayer v17.5.0. For whatever reason, PrusaSlicer seems to most often aggravate the issue on launch, but it can happen at other times too.

Info on environment:

• VMWarePlayer (licensed version) v17.5.0
• GPU: Intel ARC A370M / Intel Iris Xe
• Host OS: Windows 11
• Guest OS: Ubuntu 23.10

I've tried just about everything at this point:

• OpenGL rendering:
  • Seems to work stability, but there are major drawing artifacts that make 3d accelerated apps basically unusable
• Change Graphics memory size and number of vCPUs in VM
  • Seems to have no effect
• DX11 rendering
  • MksSandbox wont do DX11 rendering on my system
• Disable acceleration
  • Seems to work, but this givens unacceptable performance for the apps I need.

I have a paid license for VMWare player and really hope that VMWare can address this issue in an update soon. This makes my workflow miserable.

[Jose Fonseca]

 ,

Thanks for reporting this issue.  I've filed a bug internally.  The fact PrusaSlicer is free and open-source will hopefully enable us to reproduce the issue internally, and that's generally the quickest way to resolve this sort of issues.

It would help us if you could provide the vmware.log and mksSandbox.log files that are on the VM directory, immediately after the crash.

Could you also confirm exactly what did do when you tried "OpenGL rendering"?

[VMware Community Member]

thanks for the prompt reply.

> It would help us if you could provide the vmware.log and mksSandbox.log files that are on the VM directory, immediately after the crash.

Attached are the vmware.log, the mksSandbox.log as well as coredumps from right after a crash of launching PrusaSlicer.

> Could you also confirm exactly what did do when you tried "OpenGL rendering"?

I added the following snippet to my VMX file:

mks.enableDX11Renderer = "FALSE"
mks.enableDX12Renderer = "FALSE"
mks.enableGLRenderer = "TRUE"

I have confirmed that MksSandbox is using the OpenGL renderer by looking at the logs with the above settings.

Also attached is a screen shot showing some of the OpenGL rendering artifacts. It often doesn't draw everything depending on the mouse position or the position of other windows.

If there is anything else I can do to help your search let me know. I've been a VMWare user for many years, and I'll be very grateful if you can help me get my workflow back to how smooth it was in the past.

One side question: I've tried to get DX11 rending working thinking that might be more stable. However, when I disable DX12, MksSandbox tries to spin out DX11 but claims there is no compatible interface. Am I barking up the wrong tree to try this out as a work around?

Edit:

One other note: the guest OS is actually XUbuntu 23.10, not plain Ubuntu 23.10. I don't think this should make a difference, but XUbuntu still uses X11/Xorg where Ubuntu uses Wayland. Again, I don't think that should be an issue here as I've seen this type of crash on Ubuntu as well, but thought I'd point it out since the display server is interacting with the accelerated graphics.

Attachment(s)

vmmcores.zip   10.86 MB 1 version

vmware-vmx.zip   3.87 MB 1 version

vmware.log   331 KB 1 version

mksSandbox.log   113 KB 1 version

[Jose Fonseca]

 ,

Thank for the logs.  These are helpful.

> I've tried to get DX11 rending working thinking that might be more stable. However, when I disable DX12, MksSandbox tries to spin out DX11 but claims there is no compatible interface. Am I barking up the wrong tree to try this out as a work around?

DX11 renderer is still around, but there's an accept/deny-list restricting its use to older GPUs.  For two reasons: 1) because we want to eventually transition away from it,  2) because we run troubles when using DX11 renderer with new GPUs.  In fact, I believe we ran into troubles precisely with newer Intel GPUs like the you have.  So generally, DX11 is not a sustainable solution going forward, particularly for the HW you have.

OpenGL renderer is also around, but it's also mostly for diagnosis purposes, as it too has many issues, particularly on Windows hosts and Intel HW.  DX12 and Vulkan renderer are the ones we're focusing on, and we are keen on addressing any issues.

 

BTW, from the crash message, the issue looks like it could be coming from the Intel GPU crashing/resetting somehow.  Your system has two GPUs, and the crash happened with the discrete Arc GPU.  You could see if switching to the internal Xe GPU helps, by setting `mks.dx12.gpuPreference=MinPower` setting to the .vmx.

 

One more thing, could you please point us to a freely available 3D model we can use PrusaSlicer to repro this, and explain which operations trigger the crash?

 

[VMware Community Member]

 

Thanks for the info on DX11 and OpenGL. It looks like the only way out of this one is through it. Is Vulcan an option on Windows? I'm less familiar with that so don't really know if that's a silly question or not.

No model is needed to reproduce the problem in PrusaSlicer. It doesn't crash every time, but if it does it will do so before the splash screen goes away so you don't need a model to load. I'll work on a minimal reproduction steps with a fresh VM on Ubuntu and specific steps and versions of software to help you debug this. It'll take me a couple days to get time for that, but I'll post back here when I get it.

I'll also try out targeting the Iris Xe and see how that goes. Obviously performance wont match the Arc, but it'll be interesting if it doesn't have this issue. I'll also report back on how that goes.

Last note, this is top notch customer support. I didn't expect such detailed engagement directly from the engineering team. I truly appropriate it, and whether or not we crush this bug it makes me want to stay a VMWare customer into the future.

[Jose Fonseca]

,

Thanks for the updates.  

> Last note, this is top notch customer support. I didn't expect such detailed engagement directly from the engineering team. I truly appropriate it, and whether or not we crush this bug it makes me want to stay a VMWare customer into the future.

 I'm glad to hear it.  Our team is monitoring these Workstation/Fusion forums.  Not only we like to do good by all our customers, but the Fusion and Workstation users in particular provide us an exposure to a wider range of 3D applications than we could ever discover by ourselves, and therefore fixing these issues and incorporating the applications on our test suites enables to steadily improve the quality across all our products.  3D graphics APIs are vast and complex interfaces, so test coverage is the only road for quality, this applies not only to our drivers, but also the 3rd party HW vendor drivers too.  Unfortunately not all issues reported on these forums are easily reproducible and things sometimes do fall through the cracks.

My colleague,  will take it from here as I'm not directly working on the component in question.

[VMware Community Member]

and  

Thanks again for great support. I have been running with DX11 rendering now since the last post and have not had a single crash (feels so good). I'm still trying to find the time to get a minimal repro example and will post that here when I get it. In the meantime, let me know if there is any other info I can provide to help with your debugging.

 

[VMware Community Member]

  I had a chance to confirm a minimal repo on this issue. below are the steps I took:

* Install Ubuntu 23.10.1
* Defaults for everything in the VM creation (VMX included in attached zip)
* Defaults in the entire Ubuntu install process
* boot up and login
* Install prusa-slicer
  `sudo snap install prusa-slicer`
  At time of writing this was 2.7.1
* Launch prusa slicer from the app launcher
* Mine crashed on the first try at startup of the app (after a couple clicks about SSL configuration and skipping 3d printer setup)
  * If it doesn't crash on first try, try launching more instances of prusa slicer in rapid sucession without closing them. This usually triggers a crash for me within 3 instances.

Attached ate the VMX, and logs to help.

I'm on a Dell XPS 9530 with an Intel ARC A370M.

Let me know if there is anything else I can do to help your debugging and thanks again for looking into this. Also, I've had zero crashes with the DX11 render since I last posted for reference.

Attachment(s)

vmware-info.zip   7.95 MB 1 version

[Michal Krol]

 Thank you. I will try and reproduce the issue on my local system.

[VMware Community Member]

 

I just tried `mks.dx12.gpuPreference=MinPower` and confirmed that MksSandbox does indeed select the Iris Xe. Interestingly I still got the crash on my first try launching PrusaSlicer. Logs and dumps attached.

Attachment(s)

vmware-vmx.dmp.zip   2.05 MB 1 version

vmware.log   329 KB 1 version

vmmcores.gz.zip   9.74 MB 1 version

mksSandbox.log   113 KB 1 version

[Michal Krol]

> I've tried to get DX11 rending working thinking that might be more stable. However, when I disable DX12, MksSandbox tries to spin out DX11 but claims there is no compatible interface. Am I barking up the wrong tree to try this out as a work around?

Yes, it's worth giving DX11Renderer a try as a workaround. Here are the VMX config options that force-enable DX11Renderer:

mks.dx11.allowUnsupportedDevices = "TRUE"
mks.enableDX12Renderer = "FALSE"

 

[VMware Community Member]

 

Those settings did allow me to use the DX11 renderer with my Arc A370M. I tried launching multiple copies of PrusaSlicer which usually gets me a crash if it's going to happen, and so far no crash. This obviously isn't a great long term solution as DX11 will be going away and there are a couple small (but very manageable) rending artifacts, but if this gets my workflow stable while a fix is in the works for DX12 I am eternally grateful.

I still need to find some time to build a minimal repro setup and will post that here when I get a chance. Thanks again to you and your team for the help. I know how hard it is chasing hard to reproduce customer issues.

[shelterx]

Using OpenGL renderer seems to work for me as a workaround when running VMWare WS 17 in Windows 11.