VMware Communities
peter_w2019
Enthusiast
Enthusiast

Continuous Problem when switching between host and guest

Dear All

I am using Vmware Fusion 11.1 and running a Windows10 guest on a 2018 MacBook Pro. I created the windows10 guest virtual machine on a MacBook Air 2015 and recently switched.

Shortly after switching to a new computer the problem started. Every time I switch between host and guest Vmware Fusion stalls and the guest is unusable while the physical Vmware process runs at 100% CPU. When I send countless CTRL-ALT-DEL commands to the guest and finally after 5-10 min task manager comes up it works again. But this is really annoying me to death. Searching the log I see the following timeout:

2019-08-08T18:46:59.602+01:00| vmx| I125: GuestRpcSendTimedOut: message to toolbox-dnd timed out.

Does anyone have an idea or had similar issues? Any thoughts are greatly appreciated.

Many thanks and regards

Peter

19 Replies
wila
Immortal
Immortal

Hi,

It looks like you have bumped into the issue debated here: https://planetvm.net/blog/?p=67185

The good news is that there's a hotfix on the way (hopefully by Monday).

As a workaround you can temporarily change your VM RAM to only use 2GB of RAM.

--

Wil

| Author of Vimalin. The virtual machine Backup app for VMware Fusion, VMware Workstation and Player |
| More info at vimalin.com | Twitter @wilva
Reply
0 Kudos
rachelsunsm
VMware Employee
VMware Employee

Hi,

Fusion 11.1.1 is just pushed live. We have a mitigation fix for the VM performance issue on macOS 10.14.6 with this release.

Root cause for the VM performance issue, is “the behavior of paging memory from an unlinked file.” Check KSC’s updates in community if you want to understand the technical details. And this one have further details: https://communities.vmware.com/message/2877617#2877617

Per our internal testing, the performance issue has been greatly improved. If you experienced performance issue with macOS 10.14.6, pls upgrade to this build and let us know if it works for you.

Thank you for your patience! We appreciate all of you working together with us to understand the root cause and figuring out the solution!

VMware Fusion Team

peter_w2019
Enthusiast
Enthusiast

Hi there

I am afraid the issue is not yet resolved. Just a couple of minutes ago I encountered the exact same issue as before again. Since yesterday I stopped using encryption and installed the latest version of Vmware Fusion 11.1.1. It worked really well until now. Again I was seeing the timeout error as mentioned in my initial post.

I enclosed the Vmware.log for review.

I am sorry to post the bad news, but it doesn't seem to be completely resolved.

Thanks and regards,

Peter

Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

Hi All,

I still having this problem. 100%CPU while getting timeouts

2019-08-12T10:07:57.606+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.118 seconds (ok)

2019-08-12T10:07:57.606+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.119 seconds (ok)

2019-08-12T10:07:57.606+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.120 seconds (ok)

2019-08-12T10:07:57.606+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.122 seconds (ok)

2019-08-12T10:07:57.606+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.120 seconds (ok)

2019-08-12T10:07:57.606+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.117 seconds (ok)

2019-08-12T10:07:57.606+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.118 seconds (ok)

2019-08-12T10:07:57.607+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.117 seconds (ok)

2019-08-12T10:07:57.607+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.116 seconds (ok)

2019-08-12T10:07:57.607+01:00| vmx| I125: scsi0:0: Command READ(10) took 28.114 seconds (ok)

2019-08

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command WRITE(10) took 4.737 seconds (ok)

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.144 seconds (ok)

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command READ(10) took 3.007 seconds (ok)

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.142 seconds (ok)

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.144 seconds (ok)

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.144 seconds (ok)

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command READ(10) took 3.011 seconds (ok)

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command WRITE(10) took 1.170 seconds (ok)

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.016 seconds (ok)

2019-08-12T10:22:57.576+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.145 seconds (ok)

2019-08-12T10:22:57.577+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.143 seconds (ok)

2019-08-12T10:22:57.577+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.143 seconds (ok)

2019-08-12T10:22:57.577+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.146 seconds (ok)

2019-08-12T10:22:57.578+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.145 seconds (ok)

2019-08-12T10:22:57.580+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.147 seconds (ok)

2019-08-12T10:22:57.581+01:00| vmx| I125: scsi0:0: Command READ(10) took 5.148 seconds (ok)

2019-08-12T10:26:57.983+01:00| vmx| I125: GuestRpcSendTimedOut: message to toolbox-dnd timed out.

2019-08-12T10:26:57.984+01:00| vmx| I125: GuestRpcSendTimedOut: message to toolbox-dnd timed out.

has anyone seen this before?

Thanks and regards,

Peter

Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

PLEASE HELP, I can't work like this!!!!

Reply
0 Kudos
wila
Immortal
Immortal

Hi,

As mentioned in an earlier post, you have to drop the virtual Machine RAM to 2GB in order to not have this issue at the moment.

Your VM's memory is still at 9GB

2019-08-12T09:18:47.086+01:00| vmx| I125: DICT              numvcpus = "4"
2019-08-12T09:18:47.086+01:00| vmx| I125: DICT  cpuid.coresPerSocket = "4"
2019-08-12T09:18:47.086+01:00| vmx| I125: DICT               memsize = "9216"

Personally I also think you have overcommitted the CPUs on your intel i7-8559U, it only has 4 physical cores (threads don't count) and you're not leaving any for the host.

Another user in the mega thread mentioned that his/her VM worked OK with 3.6GB, so you might try that, but I'm not too optimistic about it.

A VMware employee mentioned that the "Cant' use much RAM" issue will be resolved in a later update, but that will probably be weeks away.

--

Wil

| Author of Vimalin. The virtual machine Backup app for VMware Fusion, VMware Workstation and Player |
| More info at vimalin.com | Twitter @wilva
Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

hi

thanks for you reply. I will try working with less cores, but I can't change the RAM since the application I am working with has a minimum memory requirement of 8 GB. so if I set it to use only 2 gb ram I can't work either...

thanks and regards,

peter

Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

also with only 2 cores the problem persists

Reply
0 Kudos
wila
Immortal
Immortal

Hi,

re. not being able to reduce the RAM. That basically puts you between a rock and a hard place.

re. the 2 CPU's. Your main issue is the RAM problem as evidenced by the slow reads in your log file. Sorry for distracting you on that one. I would still change it, but you've got bigger fish to fry before that even comes into play.

From the earlier posts by VMware employee ksc, let me quote a bit of when the RAM problem gets triggered:

Triggers after 2GB of modified data is cached, regardless of file. Once triggered, it stays triggered as long as the file is open.

Your VM has 9GB of RAM and the VM's RAM is backed by a file. So perhaps the issue can be mitigated a bit by turning off that feature, but I'm only stabbing in the dark here.

OTOH if your VM changes a lot of data all the time (>2GB) then it wouldn't make a difference based on the above statement

I think it would be best to open a support case with VMware and see if they can help you on a user case bases.

--

Wil

| Author of Vimalin. The virtual machine Backup app for VMware Fusion, VMware Workstation and Player |
| More info at vimalin.com | Twitter @wilva
Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

Hi there

Thanks for your input. Well, I did now reduce the numbers of CPU down to 2 and reduced the vm's RAM down to 5. This makes the application I am using difficult in handling but, at least so far, I have not experienced any stalls no more. Although I am still getting errors in the log as shown below. As long as it doesn't unexpectedly hang as before, this is some improvement for me. Still it's not a real solution for me, but for the time being until there is a fix from Vmware I am trying to work like this.

2019-08-13T07:25:29.273+01:00| vcpu-1| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T07:25:29.273+01:00| vcpu-0| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T07:25:29.274+01:00| vcpu-1| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T07:25:29.274+01:00| vcpu-1| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T07:25:29.275+01:00| vcpu-1| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T07:25:29.275+01:00| vcpu-1| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

but at least the timeout issue seems to have disappeared, which seem to have caused the stalls.

Many thanks again for your continuous support!

Cheers,

Peter

Reply
0 Kudos
DaveP
Commander
Commander

I switched the vmem file off and reverted to RAM plus swap for the VM. It seems to make a significant difference and I am running VMs now without any particular issues on MacBook Pro 15" 2018 model with 16GB RAM.

Add these 2 lines to ~/Library/Preferences/VMware Fusion/config

mainmem.backing = "swap"

mainmem.useNamedFile = "FALSE"

I would be interested in if this makes a difference for you.

Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

Hi there

I did enter the two above lines into the .vmx config file. I upped the RAM to 6 GB and after a while I got again errors and stalls.

2019-08-13T19:56:06.674+01:00| vcpu-1| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.675+01:00| vcpu-1| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.676+01:00| vcpu-2| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.676+01:00| vcpu-2| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.678+01:00| vcpu-2| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.678+01:00| vcpu-2| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.679+01:00| vcpu-2| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.679+01:00| vcpu-2| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.681+01:00| vcpu-2| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.681+01:00| vcpu-2| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.682+01:00| vcpu-0| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

2019-08-13T19:56:06.682+01:00| vcpu-0| I125: SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported page 0x8

not sure if it was the right place to put the lines to. but in ~/Library/Preferences/VMware Fusion/config

there was neither a config directory nor a config file.

but anyway, still getting errors and hangs.

Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

what version of macOS are you using?

are you running 10.14.6?

Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

now I am back at 4GB RAM, which is really the least I can set, since the app I am working with is not working otherwise....

despite the changes for swap file I am getting this nasty stalls. it is sooo annoying and let's one freak out every couple of minutes

could anyone please explain, WHAT IS REALLY GOING ON AND WHY THIS IS HAPPENING?
I am getting choleric more and more everyday, since this is not a way to work.

I start asking myself why I the world, did I buy a Vmware license for... I assumed this product works.

2019-08-14T06:57:48.480+01:00| vmx| I125: GuestRpcSendTimedOut: message to toolbox-dnd timed out.

2019-08-14T06:57:59.258+01:00| vmx| I125: GuestRpcSendTimedOut: message to toolbox-dnd timed out.

2019-08-14T06:58:04.781+01:00| mks| I125: SWBWindow: Number of MKSWindows changed: 1 rendering MKSWindow(s) of total 1.

2019-08-14T06:58:06.621+01:00| vmx| I125: scsi0:0: Command READ(10) took 1.064 seconds (ok)

2019-08-14T06:58:07.696+01:00| vmx| I125: scsi0:0: Command READ(10) took 1.075 seconds (ok)

2019-08-14T06:58:09.307+01:00| vmx| I125: scsi0:0: Command READ(10) took 1.484 seconds (ok)

2019-08-14T06:58:09.307+01:00| vmx| I125: scsi0:0: Command READ(10) took 1.488 seconds (ok)

2019-08-14T06:58:09.307+01:00| vmx| I125: scsi0:0: Command READ(10) took 1.488 seconds (ok)

2019-08-14T06:58:09.307+01:00| vmx| I125: scsi0:0: Command READ(10) took 1.604 seconds (ok)

2019-08-14T06:58:09.307+01:00| vmx| I125: scsi0:0: Command READ(10) took 1.527 seconds (ok)

2019-08-14T06:58:09.307+01:00| vmx| I125: scsi0:0: Command READ(10) took 1.290 seconds (ok)

2019-08-14T06:58:09.308+01:00| vmx| I125: scsi0:0: Command READ(10) took 1.606 seconds (ok)

2019-08-14T06:58:12.677+01:00| vcpu-1| I125: TOOLS call to  failed.

2019-08-14T06:58:24.790+01:00| vmx| I125: GuestRpcSendTimedOut: message to toolbox timed out.

2019-08-14T06:58:24.790+01:00| vmx| I125: Tools: [AppStatus] Last heartbeat value 2103 (last received 8s ago)

2019-08-14T06:58:34.269+01:00| vmx| I125: GuestRpcSendTimedOut: message to toolbox-dnd timed out.

2019-08-14T06:58:36.764+01:00| vcpu-0| I125: Tools: Tools heartbeat timeout.

2019-08-14T06:58:36.764+01:00| vcpu-0| I125: Tools: Running status rpc handler: 1 => 0.

2019-08-14T06:58:36.764+01:00| vcpu-0| I125: Tools: Changing running status: 1 => 0.

2019-08-14T06:58:36.764+01:00| vcpu-0| I125: Tools: [RunningStatus] Last heartbeat value 2103 (last received 20s ago)

Reply
0 Kudos
DaveP
Commander
Commander

It's a bug Apple introduced in 10.14.6 that Apple are working on, plus Apple changed it's memory management in a minor update that has affected the way memory is managed in macOS. Apple has acknowledged the bug and VMware released 11.1.1 as a fix. However the memory management change in macOS means much more work for VMware to address what has happened, and I suspect other software vendors.

Apple has done this before changing something between the last beta version and the gold master which is released. That's possibly why VMware may have been caught out.

I am using 10.14.1 on a MacBook Pro 15" 2018 model, MacBookPro15,1.

Can you check to see if there are .vmem files being created in the VM folder please? If so the config changes have not been applied properly.

Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

Thanks for your reply. I am getting so frustrated by this! it's really hard to explain why I freak out every couple of minutes.

I can confirm that there are no .vmem files in the folder created. there is one .nvram which is only 271 kb size.

I presume the changes are correctly applied then.

if I knew what problems this last update cause, I'd have never installed it. 

there are no word to describe the frustration. this simply has to work! I have used Vmware Fusion until 3 years ago for many years and was so happy with it, but this is not good.

I really hope they come up with a fix for this rather soon.

Reply
0 Kudos
DaveP
Commander
Commander

I understand. I have been using VMware products for 20 years since version 1 on Linux, and have never really seen this level of problem before. I have been involved in software developemnt for over 30 years and I am sure VMware devs are just as frustrated, but sometimes fixes or workarounds take time. I worked on financial big data systems using in-memory databases and sometimes finding root causes of customer issues is incredibly hard.

Reply
0 Kudos
peter_w2019
Enthusiast
Enthusiast

does anyone know around here when this problem will be fixed??? sooner or later I will loose my job, simply because I can't finish my work on time.

I am really desperate to work on a version that is stable and reliable.

Thanks for any info.

p

Reply
0 Kudos
wila
Immortal
Immortal

Peter,

Can't tell you when a real fix is out, you'd need somebody from VMware for that.

Perhaps Mikero​ is able to get your more information here or in private.

However hopefully I can give you a few alternative ideas to make it more workable.

One of the things I tend to do myself is to scale out. I'm a software developer and part of my work is consulting for other software companies.

As a result I run a lot of different software. One of the more demanding types of software is database servers.

I do tend to keep those encapsulated in their own VMs, that way I can spread the load.

That -most likely- would also work here if you are able to split the load up over multiple VMs.

Another alternative would be to go back to an earlier version of macOS.

I'm lucky as I noticed that there was problem in macOS 10.14.6 before I had hit the upgrade button and can confirm that there is no performance issue in 10.14.5.

The problem is though that you have to basically wipe your host in order to go back.

If you want to follow through that then please make full copies of your VMs (with the VM shut down) to an external disk before running a reinstall.

You are welcome to use my backup software (see my post signature) for that, but in this case a full copy is just fine.

Please do not trust Time Machine with backing up your VMs, it is fine for the rest, but don't use it for VM backups (details here, if you want to know why)

There's a risk that the reinstall comes back in 10.14.6 already, but up to a week or so ago some people confirmed they came back into 10.14.5 and then postponed the update.

Note also that macOS 10.12 and macOS 10.13 do not have this issue. If you have enough space, then you might be able to install that in a separate partition or even an external disk.

https://twocanoes.com/knowledge-base/installing-earlier-versions-of-macos-from-the-recovery-partitio...

For the moment that is all I can come up with.

--

Wil

| Author of Vimalin. The virtual machine Backup app for VMware Fusion, VMware Workstation and Player |
| More info at vimalin.com | Twitter @wilva
Reply
0 Kudos