Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Status
The first post of this thread is a WikiPost and can be edited by anyone with the appropiate permissions. Your edits will be public.

Alex-Microsmeta

macrumors 6502
Jul 14, 2018
376
630
Rome
DD is bit by bit cloning, there is no better option in my opinion, it should copy 100% of the disk.


I am positive you do it right, you are copying the whole disk, not partitions, that's the way to go.
I rechecked, you do it wrong, copy the whole disk, not the APFS container.
It's a 500 GB disk, you need a target disk of at least 500 GB

To be more precise, all partitions on the source disk have to be copied, reason, there are partitions on the disk which are needed, without them your clone is not bootable.
So, you can stop now, a 500 GB disk will take most of the day to copy, depending on the transport protocol, if you have Thunderbolt it would be fast, if it's a USB 3 disk it is a lot slower.
I once cloned a disk, can't recall how big it was, it took most of the day to copy the disk.
Yes, you confirmed my doubt about selecting the right source. I will retry with a smaller volume.
 
  • Like
Reactions: TimothyR734

Wowfunhappy

macrumors 68000
Mar 12, 2019
1,747
2,090
The catch right now with ParrotGeek's patcher is that his fork of Hax.dylib is functionally similar to ASentientBot's original Hax and not Hax2 or Hax3, and it does not try to prevent volume sealing.

Excuse me for butting into a thread I haven't been following—does this mean you can actually make live changes to the system volume without rebooting? I would very much like a way to do that in general, on supported Macs.
 

ricoc90

macrumors newbie
Jun 22, 2018
27
73
Excuse me for butting into a thread I haven't been following—does this mean you can actually make live changes to the system volume without rebooting? I would very much like a way to do that in general, on supported Macs.

Some people were talking about turning off your mac at stage 3 of the installer to prevent the installer from sealing your APFS volume, allowing it to boot from the live volume instead. I just tested this on my 2012 Mac mini and it does indeed work. I can now just mount -uw / without issues.

So basically when you start the installation of Big Sur it copies over the data to the APFS Volume (stage 1), then reboots and installs macOS (stage 2). Then it reboots again but this time you'll hold cmd+v to enter verbose mode and let it run until you see a line saying something like
Code:
executing /System/Library/Filesystems/apfs.fs/Contents/Resources/apfs_sealvolume

On which point you'll turn off your mac. Then turn your mac back on holding the option/alt key to get into the boot menu and select your Big Sur volume.
 

Wowfunhappy

macrumors 68000
Mar 12, 2019
1,747
2,090
Some people were talking about turning off your mac at stage 3 of the installer to prevent the installer from sealing your APFS volume, allowing it to boot from the live volume instead. I just tested this on my 2012 Mac mini and it does indeed work. I can now just mount -uw / without issues.

So basically when you start the installation of Big Sur it copies over the data to the APFS Volume (stage 1), then reboots and installs macOS (stage 2). Then it reboots again but this time you'll hold cmd+v to enter verbose mode and let it run until you see a line saying something like
Code:
executing /System/Library/Filesystems/apfs.fs/Contents/Resources/apfs_sealvolume

On which point you'll turn off your mac. Then turn your mac back on holding the option/alt key to get into the boot menu and select your Big Sur volume.

Neat, thanks! I do hope there's eventually a better way to do this than interrupting the installer, that feels quite tenuous!
 
Last edited:
  • Like
Reactions: TimothyR734

Payne Mononymous Bachman

macrumors member
Jun 22, 2020
97
181
Any one know why when entering sudo mount -uw / why I get an error saying volume could not be mounted: Operation not permitted and failed with 77? This is also for a mid 2012 non retina macbook pro.
 
  • Like
Reactions: TimothyR734

Wowfunhappy

macrumors 68000
Mar 12, 2019
1,747
2,090
Any one know why when entering sudo mount -uw / why I get an error saying volume could not be mounted: Operation not permitted and failed with 77? This is also for a mid 2012 non retina macbook pro.
Yes—that would be because you are using the root snapshot thingamajig and would need to go through a big rigamarrole in order to edit the snapshot. If you want to just boot from the root volume instead, see posts above.
 
  • Like
Reactions: TimothyR734

TimothyR734

macrumors 68030
Apr 10, 2018
2,723
2,753
Logsden Oregon
cool feature you can change the background in Safari :)
Screen Shot 2020-07-15 at 9.05.45 PM.png
 
  • Like
Reactions: webg3 and Maclinux

Arnomacmini

macrumors newbie
Oct 19, 2019
28
26
Hello everybody, I'm with Catalina 15.5 on Macmini 5.1 (mi-2011)
Could please someone help me to install Big sur .
I did read the posts but i can't find the good solution ?
Thanks a lot.
 

Barry K. Nathan

macrumors 6502
Jul 6, 2018
387
1,145
Irvine, CA, USA
@jackluke you could also force to include the LegacyUSBInjector in the BootKernelExtensions.kc with kmutil and the "bundle-path" option:
kmutil create -n boot --boot-path /PATH/TO/BOOT.KC --kernel /PATH/TO/KERNEL --volume-root /BS/ROOT/VOL --bundle-path /PATH/TO/LegacyUSBInjector.kext
This turned out to be the secret to getting LegacyUSBInjector.kext to work! (I have not yet gotten around to trying to change the BaseSystem.dmg yet -- this is in the context of a Big Sur installation on an external USB SSD, where I'm installing Big Sur, LegacyUSBInjector.kext, other kext patches, removing the telemetry plugin on a MacBookPro8,1 then booting the drive on my MacBook6,1.)

But, the weird thing is that now my MacBookPro8,1 and MacBook6,1 are booting from different APFS snapshots on the same disk. The MacBookPro8,1 is booting from the APFS snapshot that it's supposed to, but the MacBook6,1 is booting from the snapshot that the Big Sur installer created, not any of the newer snapshots. I discovered this because the MacBook6,1 was having the usual kernel panic from the telemetry plugin, and I could see in single-user mode that the plugin was still there, but when I booted the same disk in single-user mode on the MBP8,1 I saw that the plugin was renamed as I had intended -- and both Macs were booting from snapshots! (It's not a case of one Mac booting from a snapshot and the other Mac booting straight from the APFS volume.) But it turns out they're booting from snapshots with different xids (this is visible in the boot messages). This is with beta 2, by the way.

I just finished making a dd clone of the external SSD where this is happening (I've been using dd to clone macOS betas since High Sierra). I'll investigate the situation a bit more, but I just wanted to alert everyone that this is possible. (Has anyone else had this happen yet?)
 

Alex-Microsmeta

macrumors 6502
Jul 14, 2018
376
630
Rome
macOS 10.15.6 doesn't fix Disk Utility about BS partitions ? Now, even worse, after most cloning trials, instead of Update partition + Big Sur - Data, on Internal SSD, toghether with Catalina it shows only 1 ASR container that cannot be fixed. Luckily BS beta 2 still loads flawlessy ? I'll wait beta 3 to reinstall BS in a working container.
 
  • Like
Reactions: TimothyR734

jackluke

macrumors 68040
Jun 15, 2018
3,321
8,068
This turned out to be the secret to getting LegacyUSBInjector.kext to work! (I have not yet gotten around to trying to change the BaseSystem.dmg yet -- this is in the context of a Big Sur installation on an external USB SSD, where I'm installing Big Sur, LegacyUSBInjector.kext, other kext patches, removing the telemetry plugin on a MacBookPro8,1 then booting the drive on my MacBook6,1.)

It's not a secret, you just need to use the prelinkedkernel (also the apple stock one if you prefer) as main Preboot BigSur kernelcache: com.apple.Boot.plist

this "prelinkedkernel method" is mandatory if you want to build a legacy USB prelinkedkernel, otherwise kextcache delegates its tasks to kmutil that rebuild only the BootKernelExtensions (that I can't make work with LegacyUSBInjector).

To make LegacyUSBInjector.kext to work with prelinkedkernel you need to add it to BigSur SLE (you need an unsealed system with no snapshot), but using only this will boot legacy USB but with "CMD+S" and "exit".

I instead discovered that prelinking it together with this (that should provide also SSE4.2 and AVX emulation):
https://forums.macrumors.com/thread...emulation-to-enable-amd-metal-driver.2206682/

allows that "BigSur single user mode booting delay" I needed to detect correctly HID: Legacy Shim 2 for any non-APFS or legacy USB Penryn Core2Duo mac .

I haven't made yet a new "beta2 prelinkedkernel" because BigSur kernel is still version 20.0.0 and I assume all the prelinked kext are also almost identical to the beta1 .
 

Barry K. Nathan

macrumors 6502
Jul 6, 2018
387
1,145
Irvine, CA, USA
this "prelinkedkernel method" is mandatory if you want to build a legacy USB prelinkedkernel, otherwise kextcache delegates its tasks to kmutil that rebuild only the BootKernelExtensions (that I can't make work with LegacyUSBInjector).
Let me rephrase it: What testheit posted was, for me, the secret to getting kmutil to build a BootKernelExtensions.kc with a working LegacyUSBInjector.

Actually, there are two other catches, I remember now:

1. kmutil can build both the BootKernelExtensions.kc and the SystemKernelExtensions.kc in one command, but if you do that, LegacyUSBInjector won't work. You have to build the two of them with two different commands.

2. You have to use a chroot. I ran into some weird error with kmutil that magically disappeared once I started using a chroot.

The code that I have that does it is something like this ("$VOLUME" is the path to the Big Sur system volume):
Code:
chroot "$VOLUME" kmutil create -n boot \
    --kernel /System/Library/Kernels/kernel \
    --volume-root / \
    --bundle-path /System/Library/Extensions/LegacyUSBInjector.kext \
    --boot-path /System/Library/KernelCollections/BootKernelExtensions.kc
chroot "$VOLUME" kmutil create -n sys \
    --kernel /System/Library/Kernels/kernel \
    --volume-root / \
    --system-path /System/Library/KernelCollections/SystemKernelExtensions.kc \
    --boot-path /System/Library/KernelCollections/BootKernelExtensions.kc
"$VOLUME/usr/sbin/kcditto"
bless --folder "$VOLUME"/System/Library/CoreServices --bootefi --create-snapshot
I ran this on my MacBookPro8,1 then booted the result on my MacBook6,1 and it worked. (I tried something else after that, and that's what caused my current situation where the two MacBooks boot from different snapshots on the same external SSD.)

By the way, you can run strings on the generated BootKernelExtensions.kc and look for "com.parrotgeek" and see that the LegacyUSBInjector is there (or see that it's not there if something went wrong).
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.