MatoHP's Posts

I thought it helped me but I spoke too soon. After a short while I was back in the annoying, long update loop. During my google journey I found this: https://answers.microsoft.com/en-us/windows/forum/all/installation-failure-for-2023-10-cumulative-update/7a2d7e43-363f-4415-8456-351f19cf2154 It seems something was wrong with the early Oct update coming from MS. While suggestions there didn't help me (dism rollback failed) it gave me an idea. I uninstalled all KBs applied to that system from Oct 2023. And then let the system be updated again. This did solve my problem. I'm back on nvme disks, no more update loop. I tested it for two days, few other updates were successfully installed too.

Tonight was one of those nights where I just wanted to have the VM updated and be on my way. I hit the issue with this update, VM kept downloading, rebooting, failing updates. I found nothing obvious at the first glance. Temporarily changing disk to sata from nvme solved it. Thanks.

While this is obvious to many I guess there might still be people like me who held back their setup because of this bug. I like to do a follow up on older threads I asked something. I tested it first and later did a fresh install of Catalina. Indeed it seems this bug is solved now. I've seen no issues past this week or so running several VMs doing their job. I think (subjective though) that network performance got better with newer os/fusion. @ColoradoMarmotI was little bit hesitant to go to Catalina and loose 32b support but then most of the things I do involve qemu/fusion where I don't have problems. I'm conservative though, I didn't jump to BigSur. But I bought Fusion 12;-)

So we have Mar 2021 now. Do you guys know if the issue is solved now ? As I mentioned in 2019 the only thing working for me was downgrade. And Apple hates that a lot. I was lucky enough to have the High Sierra VM and hence was able to download the installation files to actually do the downgrade. I'm on 10.13.6 (17G14042) , Fusion 10.1.6 (12989998). I was thinking about upgrading to Mojave. This thread is dead now, not sure if people solved it or let it be.

I really didn't have time to go through the VMware tech support ( at work we use ESXs as may corporations do -- it was hell and in the end they didn't help at all ). And we pay lots of money there. With single user license my hopes are really small to none for them to solve this issue (which is probably more complex). This is why I chose to rollback both OSX to High Sierra (10.13.6) and VMware fusion (10.1.5) and I have no problems. For anyone who really needs VMware@osx it's really the path of least resistance.

It was the worst with windows 7 guests. They got so unusable I had to kill them from cmd line. I had to uninstall them first because fusion 10.x refused to do a reinstall. I use open-vm-tools where possible.   All my linux/freebsd/.. VMs are without X , there I didn't notice problems. But I didn't test too deep either -- I had to have windows7 clients back. But once windows VMs started to have problems so did osx. I'm using these VMs for few hours now and so far so good, no problems.

I downgraded osx back to 10.13.6 (17G3025), fusion 10.1.3 (9472307). It was somewhat better but something was still off (performance was not as before). I remembered that I did upgrade vmware tools already on some machines. I removed the tools from those guests and installed new ones (10.x one). So far performance is night and day -- all guests are responding swiftly, the way they should. I'll update this thread later once those VMs have some uptime. If somebody is out there who is still on Mojave but downgraded to fusion 10.x it might be worth trying to remove/install vmtools, just to see if that helps. Of course only if they were upgraded on fusion 11.x. I didn't try to  disable Metal and enable back OpenGL as ITdojo suggested -- I was already set for downgrade.

I downgraded fusion to 10.1.3 (9472307). What do you know, there was an update out to Mojave too (though I think it's more security related to recent face time issues and others..). I know I shouldn't change two things at the same time. But I went for it, I'm more-less convinced I need to downgrade osx anyway.  So I've updated Mojave to 10.14.1 (18B75). I see some improvements now -- performance is definitely less terrible. Response time of those VMs is better, I could use them. But still, it's not the same as before. I'll update once I've reinstall osx.

I've the same issue as OP described, mostly noticeable on Windows 7 guests. Setup is Mojave 10.14 and fusion  11.0.0 (10120384). Similarly I've upgraded both fusion and osx close to each other so I can't really tell. I wonder if it has something to do with CVE-2018-5407 too. I'm planning to downgrade fusion first and then osx. I will update this thread with my findings. I just wanted to say I've the same (very annoying) problem.

ok, finally i was able to check it directly .. actually, both hints are useful - i can set the lines with mode to 25 in fullscreen and correct them with 'Stretch Guest' screen option to make a perfect fit. regarding the screen lagging - i'm using inline assembler there and due to my testing ntdvm.exe was still running after BC was closed taking all CPU resources. so thanks a lot for tips, sure did help

hi mdunn, thanks for the tip - that's a good one. It helps - when i set the lines to 25 i got the `original' full-screen size. It helps for BC too, but i see that screen is lagging - it takes much more to refresh /redraw the screen than the original size. Right now i'm connecting via rdesktop to the host where workstation is running, hence i can't confirm if that is the NW lag or not (it works ok with the original resolution though). I will be able to confirm it this tuesday. thanks!

i found out one way to deal with this - to set the smaller resolution in winxp .. it's not pretty (16:10 vs 4:3) but definitely does the trick for Borland C++ not the perfect solution, but at least something :smileycool: