macOS High Sierra - Final Release is out! (17A365)

[xtraa]

RX480, i5-HD530 and GA-Z170 fully working with Sierra modifications and apfs.

 

After digging through some threads I was preparing for a painful upgrade. Instead, it turned out to be one of the smoothest transitions ever.

 

I started with an external testdrive and installed on apfs to see if Clover is ok with the new filesystem. After that I updated my existing Sierra.  I'm using Clover v.4128 with GM apfs.efi driver, own dsdt & ssdt and same patches as in Sierra.

 

Just started the inplace update and boom, everything works like a charm. I really did not expect that. The RX480 is detected correctly as a Radeon Pro 480/575 and my idle Intel HD Graphics 530 also supports Metal. Recovery also boots fine and the "Preboot Filevault" does too, even if I have to take a deeper look in here. (only have an idea what it's purpose is, yea filevault obviously).

[xtraa]

Something has to be wrong with my system.

I see so many people getting the Update done on the fly.

Of course for some stupid reason it always doesn't work for me.

I just don't get it. My system is broken now too. Only way to start it now is with the 

-x command. Damn

 

I think we've all been at that point a few times. This happens when you do it right. (instead of doing it wrong and use some ready to go thing that will patch your system with loads of modded kext files but you don't know exactly what's going on and for the price of what)

 

If you can get to the desktop with -x it's not that bad. Without knowing your system I'd recommend this: In most cases it's the best option to start excluding the graphics drivers first. You are booting VESA mode anyway, so I'd recommend to start with an mkdir an make a new folder called e.g. extensions_disabled and move your graphics drivers there from S/L/E. From there you can see if it boots fine w/o -x and start adding drivers back. You can always move them back by booting into -s, doing fsck and mount and after that just cd .. in your directory and move files with mv. 

[SunKi]

Fix for ICH10 with High Sierra. Now with hot-plug support.

 

For this we need two binary patches.

First patch fix I/O error for ICH10:

4585F60F 95C289C8 83E0FE66 85C9780F 84D2750B -> 89C883E0 FE6685C9 0F98C141 08CC9090 9090750B 

With second patch will work hot-plug:

89C82540 6002003D 40200000 750C81E1 BF7FFFFF 898B5101 0000 -> 90909090 90909090 90909090 90909090 90909090 90909090 9090

With these two patches I got ICH10 with hot-plug.

 

Good luck

[arsradu]

No it doesn't work anymore, only createinstallmedia or startosinstall method works.

 

as @camillionario stated on some systems having the EmuVaraiableUEFI.efi can also cause the OSInstall.mpkg error.

 

Actually, with my last installation of the final build of 10.13, I didn't remove the EmuVariableUEFI.efi anymore. I upgraded straight from 10.12.6 to 10.13. And everything went smoothly. So this might have been a bug on Apple's side. But personally, I haven't encountered it anymore. Hope it won't come back hunting me in the future.

Hi,

 

i'm having problems currently on updating from the macOS High Sierra Beta to High Sierra Final. Tried everything, OTA Update, Boot via USB Flash Drive. Just nothing works.

I'm always stuck in the loading screen and the picture is just frozen with the Apple Logo and the loading circle. 

 

I have to say that i had some of these crazy problems the time i installed the beta too. I used the EFI folder from @cyberdevs and got it working. After updating i put back my EFI folder and everything worked fine.

Maybe i have the same problem again? I already updated the SMBIOS.

 

I'm currently using these SMBIOS settings:

 

BiosVersion: MP61.88Z.0120.B00.1708080652

Release Date: 08/08/2017

Firmware Features: 0xE80FE136

Features Mask: 0xFF1FFF3F

Plattform Features: 0x04

 

Maybe someone is nice enough to give me his EFI folder so i can try to get it work that way?

 

That looks like a Clover/EFI issue to me. What did it change in there from Beta to Final? I remember having this issue while trying to recover a Skylake build. In my case, it was just missing kexts in Clover/kexts/Other. I just forgot to add the kexts (especially FakeSMC) in there. I managed to get Clover installed on the USB drive, using another computer, but I totally forgot to add the kexts, too. So i was having all kinds of crazy issue.

 

By the way, I understand your frustration. I had to recover a Skylake build after simply updating it to 10.13 and installing the Nvidia drivers. And now that I think about it, there might have been a better way to fix it. But you know, when you're in the moment, you don't always find the best solution. So you try what you know hoping that it will work. Sometime does, sometimes (like in my case booting with nv_disable=1), doesn't. So you need to improvise.

 

Like creating a bootable USB with Clover on it. Having all the right files on it. Optionally with MacOS install on it, as well. But Clover is essential here.

 

And, pro tip: update your kexts once in a while. Not having NvidiaGraphicsFixup.kext updated from 1.1.2 to 1.1.3 (if you think that's a minor update, well, you would be wrong) cost me a lot of time and stress. So yeah, update your kexts once in a while.

 

And pro tip #2: try to use, as much as possible, supported hardware. Do your research before trying to make a computer run software that's not supposed to run by default. Having a supported video card, for example, can save you from A TON of pain and frustration. And this is just one example. So do your research beforehand.

[nmano]

DVD player not work

 

There was an initialization error

The operation is not currently supported.[-70014]

 

If any idea how to fix this

Thank you.

[Trung_Nguyen]

My HS read any drive from USB2.0 port as internal even patched AppleAHCIPort or not but with the USB3.0 port it is ok. How to fix this?

 

The Patch interface:

AppleAHCIPort Find 45787465 726e616c Replace 496e7465 726e616c 

<key>Comment</key>

<string>External icons patch</string>

<key>Disabled</key>

<false/>

<key>Find</key>

<data>

RXh0ZXJuYWw=

</data>

<key>Name</key>

<string>AppleAHCIPort</string>

<key>Replace</key>

<data>

SW50ZXJuYWw=

</data>

[Matgen84]

What about 7-series and asmedia? Did it work with the 328 version? Have you tried them with the patch?

 

Yes, hot-plug is working. No, it does not work .. I need to think .. Everything works finally

Sorry for my bad english. Do you say that your patches work with applaAHCI v329 for 7-series and asmediia (Hot-Plug). Right?

 

For instant, I use modded v328 by fusion71au only for Hot-Plug.

[Cyberdevs]

Actually, with my last installation of the final build of 10.13, I didn't remove the EmuVariableUEFI.efi anymore. I upgraded straight from 10.12.6 to 10.13. And everything went smoothly. So this might have been a bug on Apple's side. But personally, I haven't encountered it anymore. Hope it won't come back hunting me in the future.

I have successfully installed macOS High Sierra on my SkyLake rig with EmuVariableUEFI.efi present in my drivers64uefi folder as well and didn't encounter the notorious OSInstall.mpkg error. But it appears that it happened to @SavageAUS while he was trying to install macOS High Sierra, Me and SavageAUS have almost the same hardware with two different GPUs. So I'm not sure why on some cases it is still happening.

 

I haven't tried the installing High Sierra on my Asus rig with the EmuVariableUEFI.efi yet but the OSInstall.mpkg error was present up until the latest beta version, so I don't use the EmuVariableUEFI.efi for two reasons:

 

1. I don't need the EMuVariableUEFI

2. It was giving me the OSInstall.mpkg error.

 

But If I find the time I will give it a try and let you know if it's gone forever or is it still going to happen on some hardware.

[SunKi]

Do you say that your patches work with applaAHCI v329 for 7-series and asmediia (Hot-Plug). Right?

Just try it and tell us. I do not have this hardware.. Patches from this post - http://www.insanelymac.com/forum/topic/327488-macos-high-sierra-final-release-is-out-17a365/?p=2505086

[Guest ricoc90]

I did not encounter the OSinstall.mpkg error since DP2 up to the final release.

However, It happened to me when installing 10.13.1 beta 1. I'm on legacy ESP install, so no EmuVariableUEFI.efi. Disconnected all drives except the one I was installing too (...), but still ran into this error.

Removed install data, redownloaded the update, ran nvram -c and was able to run the second stage without problems.

[Badruzeus]

On my laptop with already installed Windows & Ubuntu on another partitions (Single Legacy-GPT HDD) however, I have no idea how to choose target volume for installing if plugged it out. Since DP1 till 10.13.1 Beta, I just unchecked rc.script on Clover install and it did the trick, removed nvram.plist, uhmmmb.. for sure, with updated platform datas within bootloader. But stange is on another one with similar specs and multiple OSes on single HDD, same clover rev, same macOS version it just doesn't work and got OSInstall.mpkg error..

[Matgen84]

Just try it and tell us. I do not have this hardware.. Patches from this post - http://www.insanelymac.com/forum/topic/327488-macos-high-sierra-final-release-is-out-17a365/?p=2505086

 

 

Because all the sata ports are well recognized, I try only Hot-Plug patch : Hot-Pluging work like a charm on my system. Great Job 

[Badruzeus]

Firmware error with High Sierra (real mac)

[chris1111]

Because all the sata ports are well recognized, I try only Hot-Plug patch : Hot-Pluging work like a charm on my system. Great Job 

So you not need to rollBack old kext?

[Cyberdevs]

I have had also that OSInstall.mpkg error but in my case was related to multiple efi partition in different HDDs.

disabling one hd I can install in every mode (from usb pen or from a working osx installation)

 

I did not encounter the OSinstall.mpkg error since DP2 up to the final release.

However, It happened to me when installing 10.13.1 beta 1. I'm on legacy ESP install, so no EmuVariableUEFI.efi. Disconnected all drives except the one I was installing too (...), but still ran into this error.

Removed install data, redownloaded the update, ran nvram -c and was able to run the second stage without problems.

 

 

On my laptop with already installed Windows & Ubuntu on another partitions (Single Legacy-GPT HDD) however, I have no idea how to choose target volume for installing if plugged it out. Since DP1 till 10.13.1 Beta, I just unchecked rc.script on Clover install and it did the trick, removed nvram.plist, uhmmmb.. for sure, with updated platform datas within bootloader. But stange is on another one with similar specs and multiple OSes on single HDD, same clover rev, same macOS version it just doesn't work and got OSInstall.mpkg error..

 

I have multiple HDD's with multiple EFI partitions on my both systems. 

 

From what I see it was related to both EmuVariableUEFI and the FFM & FF and with using the latest BIOS Version and BIOS Release Date in the config.plist it could be avoided.

 

My Asus rig didn't like the EmuVariableUEFI and it wasn't needed it anyway but the Gigabyte rig requires it for the nVidia GPU to be operational. I guess with implementing the correct FFM and the FF (Thanks to @Sherlocks and developers of the clover) I believe it has been resolved in the latest clover revisions.

 

But the only thing that is certain is that on many cases the error happens for different reasons. So it's not a definite answer to why it is happening and what could be the cause for the OSInstall.mpkg error.

[Matgen84]

So you not need to rollBack old kext?

With hot-plug patch on v329, modded AppleAHCIPort.kext (L/E) or rollback old kext v328 isn't necessary. I tested it on the GM but it will be the same on the release!

[Guest ricoc90]

@cyberdevs:

 

My FFM and FF are always up to date (that is, the values that are implemented within Clover), so that shouldn't have been the problem in my case... 

By the way: updating my ProBook (Clover UEFI, EmuVariableUEFI.efi present) went really smooth. 

 

At the same time, macrumors is full of people having the exact same problem on a real mac and as far as I can see it's mostly due to a corrupted EFI partition or third party harddrive/SSD.

 

Weird.

[Cyberdevs]

@cyberdevs:

 

My FFM and FF are always up to date (that is, the values that are implemented within Clover), so that shouldn't have been the problem in my case... 

By the way: updating my ProBook (Clover UEFI, EmuVariableUEFI.efi present) went really smooth. 

 

At the same time, macrumors is full of people having the exact same problem on a real mac and as far as I can see it's mostly due to a corrupted EFI partition or third party harddrive/SSD.

 

Weird.

Yes that is weird indeed. I had some issues with multiple HDDs being attached to the system a while back and I had to disconnect them all to avoid the error. That was before I discovered the role that the EmuVariableUEFI.efi and the nvram.plist were playing (although nvram.plist never caused the error on my setups but some other people claimed that removing it helped).

 

Anyways I will be doing some more tests later on to see if it's still happening to me or not and if so why.

[Guest ricoc90]

Yup. Well like I said clearing NVRAM (sudo nvram -c) did the trick for me, so I can see that nvram.plist could be an issue. We'll see how it goes when 10.13.1 beta 2 is there 

[gpatpandp]

I did not encounter the OSinstall.mpkg error since DP2 up to the final release.

However, It happened to me when installing 10.13.1 beta 1. I'm on legacy ESP install, so no EmuVariableUEFI.efi. Disconnected all drives except the one I was installing too (...), but still ran into this error.

Removed install data, redownloaded the update, ran nvram -c and was able to run the second stage without problems.

 

I too experienced this exact same issue when attempting to update a legacy ESP system to the 10.13.1 Beta 1 update directly from the App Store.   Could you list any/all terminal commands (is it just: sudo nvram -c ?) you used to delete the nvram files from your system which then allowed you to successfully apply the update ?   Thank you !

[MICKHAEL]

Fix for ICH10 with High Sierra. Now with hot-plug support.

 

For this we need two binary patches.

First patch fix I/O error for ICH10:

4585F60F 95C289C8 83E0FE66 85C9780F 84D2750B -> 89C883E0 FE6685C9 0F98C141 08CC9090 9090750B 

With second patch will work hot-plug:

89C82540 6002003D 40200000 750C81E1 BF7FFFFF 898B5101 0000 -> 90909090 90909090 90909090 90909090 90909090 90909090 9090

With these two patches I got ICH10 with hot-plug.

Снимок экрана 2017-09-30 в 14.40.42.png

 

Good luck

hello. hotpatch in ICH9 not working.

[Guest ricoc90]

I too experienced this exact same issue when attempting to update a legacy ESP system to the 10.13.1 Beta 1 update directly from the App Store.   Could you list any/all terminal commands (is it just: sudo nvram -c ?) you used to delete the nvram files from your system which then allowed you to successfully apply the update ?   Thank you !

I removed the install data (/macOS Install Data/) folder, cleared NVRAM by indeed "sudo nvram -c" and redownloaded the update again.

I did not delete nvram.plist

[KGP-iMacPro]

Carbon Copy Cloner (CCC) now able to successfully and directly clone a macOS High Sierra 10.13 APSF Drive! 

 

Straight system disk backup also for APFS Drives! Gorgeous!    

[Guest ricoc90]

Carbon Copy Cloner (CCC) now able to successfully and directly clone a macOS High Sierra 10.13 APSF Drive! 

 

Straight system disk backup also for APFS Drives! Gorgeous!    

 

Yeah, I was stupid enough to install High Sierra on case-sensitive APFS. Used CCC to clone to APFS (non-sensitive), worked great

[Apple Hunter]

Today I wanna update EL Capitan to High Sierra So I downloaded Install macOS High Sierra.app from App Store instead of showing 5GB file Show 14MB. I delete that file from Application even from recycle bin and re download from app store still same no difference..

Now how to create USB installer ??