Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
MP 1,1-5,1 Mac OS 11.3 has broken support for older Mac Pros
- Thread starter VitaminK
- WikiPost WikiPost
- Start date
- Sort by reaction score
- Status
- Status
I've started my reboot tests, and completed 5 successful reboots after finally getting through the installation. The 6th attempt froze after initializing my Lenovo wired mouse, connected to an Apple wired keyboard (so both mouse & keyboard initialized, screen froze, then a minute later it turned to garbage with the prohibited symbol).
After powering off, the 6th boot attempt was successful. The 7th just froze while initializing the ethernet controllers (which it's done many times at that point throughout the betas). I'll do a few more reboots, and open to test suggestions.
I tried 10 reboots, without issues. @startergo @cdf @tsialex I updated #300 with what I identified as root cause on my setup, I can replicate it every time consistently. In my case, I might have better success because the chip used into Syba SI-PEX40129, which is the same as HighPoint 7101A.
Last edited:
I've seen the "you shut down your computer because of a problem" message plenty of times, however with this clean 11.3 installation, I'm not even logging in and it randomly succeeds, randomly fails. I'm simply clicking "Restart" on the login screen, over and over.
I just got a fail at:
getVolumeGroupMountFrom:7212: disk6s2 nx_get_volume_group(######) failed with error 2
Next boot was successful, logged in to see which disk that is, and it's it's the physical SATA SSD on PCIe card I was booting from, "Apple_APFS Container disk7".
getVolumeGroupMountFrom:7212: disk6s2 nx_get_volume_group(######) failed with error 2
Next boot was successful, logged in to see which disk that is, and it's it's the physical SATA SSD on PCIe card I was booting from, "Apple_APFS Container disk7".
Last edited:
I can't even boot right now.
It's weird, every time the crash is a different point, I also got a crash with the AppleBCMWLANCoreMac kext. I'll probably have to strip down everything but the GPU to at least have some consistency.
I got that one too. That is why I wanted to update the IO80211 kexts.
In my case, I can successfully restart many times, without any issues (I'm presented every time with the "you shut down your computer because of a problem." error). As soon as I do a regular shut down, I'm stuck at the Proress Bar only once and after a forced shut down, I can start my Mac as usual.
This is interesting. Maybe your configuration somehow mitigates the apparent race condition on warm boots. Now the question is, how many can you do in a row?
Anyone has tried to install macos with nvme from other hack/real mac and then plug that nvme into mac pro?
I tried your method, and got through 3 iterations of successful boot, login, error message, restart, and then it hung at yet another random place. So, your theory/method does not work in my fairly simple TEST system. I'm curious why it does work in your system.
PS - I just tried with nothing plugged in but power and DVI (no keyboard, mouse, network, etc.) and it hung after the 2nd iteration (again on the ethernet adapters - a common fail point).
Last edited:
It looks like 11.3 is the same build As RC 20E232. Why are we expecting different results?
I am such a bonehead, totally forgot my Mac Pro 5,1 was running open core and updated to 11.3 today. Like many of you my update screen progress bar frozen, and now when I boot the machine back up it takes me to an Install Screen for Mojave. I was running Big Sur off of an NVME connected with a PCI-E drive and Open core installed on an SSD inside the machine.
Thankfully I have 2 backups clones from earlier today plus the NVME drive so I think from a data standpoint I am okay.
I just can't select a different start-up disc, or re-store from either of my Time Machine backups to even get the system back up and running with 11.2. I am happy that this isn't my main machine for work as all of the OpneCore guides suggest but its my favorite machine I'd love to get it up and running again.
Since it's late I didn't want to dive too far into any real troubleshooting but if anyone knows of a guide or site that can help me roll this back to at least be able to use the machine you'll be my fav person on the web!
Thankfully I have 2 backups clones from earlier today plus the NVME drive so I think from a data standpoint I am okay.
I just can't select a different start-up disc, or re-store from either of my Time Machine backups to even get the system back up and running with 11.2. I am happy that this isn't my main machine for work as all of the OpneCore guides suggest but its my favorite machine I'd love to get it up and running again.
Since it's late I didn't want to dive too far into any real troubleshooting but if anyone knows of a guide or site that can help me roll this back to at least be able to use the machine you'll be my fav person on the web!
I use Carbon Copy Cloner to make my bootable backups. Just clone your bootable backup, back to your system drive.
I don't suppose this will really solve anything, but perhaps the experts should be aware that updating 11.2.3 to 11.3 on a Parallels virtual machine fails repeatedly. I've tried twice.
The first time, I let Software Update do the installation. After a kernel panic, Parallels returned to 11.2.3 on its own and that was that.
The second time, I downloaded the 11.3 installer from the App Store and then ran it. It took longer to install (sort of), but the end result was the same: kernel panic and then Parallels went back to 11.2.3 on its own.
I'll try creating a new 11.3 virtual machine from scratch and see if it is any more successful.
If the above experience is truly indicative of anything, I would think the issue on some physical Macs (like the cMP 5,1) and some virtual machines (like a Parallels 16.1.2 virtual machine of Big Sur) is not primarily related to NVMe or PCI (I very much doubt a Parallels virtual machine has any PCI components).
The first time, I let Software Update do the installation. After a kernel panic, Parallels returned to 11.2.3 on its own and that was that.
The second time, I downloaded the 11.3 installer from the App Store and then ran it. It took longer to install (sort of), but the end result was the same: kernel panic and then Parallels went back to 11.2.3 on its own.
I'll try creating a new 11.3 virtual machine from scratch and see if it is any more successful.
If the above experience is truly indicative of anything, I would think the issue on some physical Macs (like the cMP 5,1) and some virtual machines (like a Parallels 16.1.2 virtual machine of Big Sur) is not primarily related to NVMe or PCI (I very much doubt a Parallels virtual machine has any PCI components).
Of course it could. If so, perhaps the solution for the cMP might be similar to whatever Parallels programmers are doing or have done to solve the problem.
Does anyone have any additional info about the "You shut down your computer because of a problem" issue? The fact that it's also linked to whether or not updates work without stalling - apparently, good research @TECK ! - makes it sound definitely worth addressing - within OC or wherever - if possible.
I can confirm that I get the same error a lot on a MacBookPro10,2 (so it's definitely not just a MacPro issue).
I'm not 100% sure if the error is caused by OC or not (e.g. my MBP can boot and run BS just fine without OC, just with NVRAM "boot-args=-no_compat_check"; it just won't fetch or install updates without OC), but I haven't tested enough to be sure if the error happens anyway without OC.
I also don't currently know where the error condition is stored (by Apple; is it NVRAM?), nor what (more exactly) normally triggers it to be stored - if anyone can help with any additional info on that?
- Status