macOS 15 Sequoia on Unsupported Macs Thread

[ronton3]

I thought 2.4/15b3 was so perfect----when an email is clicked there is nothing but the heading, no content, and I can't type on an opened email to reply, is this a known problem, how to fix it. Thanks.

 

[hvds]

I upgraded from Sequoia 15.5 Beta 2 -> Beta 3 in a test volume on my HackBookPro6,2. The upgrade was smooth, but Safaria 18.5 in Beta 3 is no longer able to load MacRumors web pages (was working in Beta 2). Firefox works fine.

The good news is that I no longer have the vertical scroll bar issue in Safari (since I can't load any pages).

MBP5,2: nothing new, just to confirm these Safari findings.
Macrumors pages work fine with 15.4/Safari 18.4 but fail with 15.5b3/Safari 18.5. Writing this from Firefox.
BTW Safari 18.5 failed in 15.5b3 also before applying root patches.

Both installations done from USB installers with OCLP 2.4.0n.
KDK for 15.4 (24E248) is used in both cases by root patching.

Still amazed how usable this 2009 system is with Sequoia 15.5b and OCLP... Thanks!
i.e. WLAN, internal USB1.1 devices and graphics acceleration work.

 

[mib01]

When I upgraded from Ventura to Sequoia, it was not necessary to apply the post-install. However, when I upgraded via OTA from 15.4 to 15.4.1, the KDK was downloaded first and then I had to run the post-install.

Now comes the curiosity, were the Kexts kept in this update from Ventura to Sequoia?

 

[Neon Ball]

When I upgraded from Ventura to Sequoia, it was not necessary to apply the post-install. However, when I upgraded via OTA from 15.4 to 15.4.1, the KDK was downloaded first and then I had to run the post-install.

Now comes the curiosity, were the Kexts kept in this update from Ventura to Sequoia?

Upgrading/updating via USB drive it gets autopatched with patches on USB drive, updating via OTA it does not.

 

[Larsvonhier]

Strange OCLP behaviour on our MP4,1 / MP5,1 machines with recent combo of Sequoia 15.4.x and OCLP 2.3.2:

Install/Update of macOS works faultlessly, patching for Kepler GPU and vintage BT/Wifi/USB also.
OC loader configured as always/previously, installed to EFI boot volume.

Now the machines in this configuration boot a couple of times with everything solid, patches ok and normal behaviour.
After a while (couple of boots after proper shutdowns) the systems show unpatched (GPU/Kepler) behaviour - no acceleration, only one finder screen on one or two TFT panels (mirrored instead of extended).

Re-patching sometimes crashes (freezes the whole system shortly after NVDA extensions are re-copied), sometimes works directly on first attempt. Second try (after another reboot) always completes successfully - and "holds" again for a couple of boot-ups. Then again "magically" reverts back to unpatched Nvidia Kepler performance.

We now found out on our office machines that going back to OCLP 2.3.0 patches (while keeping latest set of OC EFI stuff on boot volume) seems to be a cure. We booted several affected machines now for a couple of times without patches "degrading".

The mechanics behind this degradation is unclear though- I could only hint at broken volume sealing that reverts back to last valid snapshot?

Hope this helps other users affected by this and perhaps leads to a remedy in upcoming OCLP versions 2.x.x !

 

[Neon Ball]

Strange OCLP behaviour on our MP4,1 / MP5,1 machines with recent combo of Sequoia 15.4.x and OCLP 2.3.2:

Install/Update of macOS works faultlessly, patching for Kepler GPU and vintage BT/Wifi/USB also.
OC loader configured as always/previously, installed to EFI boot volume.

Now the machines in this configuration boot a couple of times with everything solid, patches ok and normal behaviour.
After a while (couple of boots after proper shutdowns) the systems show unpatched (GPU/Kepler) behaviour - no acceleration, only one finder screen on one or two TFT panels (mirrored instead of extended).

Re-patching sometimes crashes (freezes the whole system shortly after NVDA extensions are re-copied), sometimes works directly on first attempt. Second try (after another reboot) always completes successfully - and "holds" again for a couple of boot-ups. Then again "magically" reverts back to unpatched Nvidia Kepler performance.

We now found out on our office machines that going back to OCLP 2.3.0 patches (while keeping latest set of OC EFI stuff on boot volume) seems to be a cure. We booted several affected machines now for a couple of times without patches "degrading".

The mechanics behind this degradation is unclear though- I could only hint at broken volume sealing that reverts back to last valid snapshot?

Hope this helps other users affected by this and perhaps leads to a remedy in upcoming OCLP versions 2.x.x !

Make sure to disable all automatic macOS updates (all the way to downloading) as they can now modify the system volume while staging the update (which reseals the snapshot), which can lead to broken patches. Sequoia 15.4 recently prompted to enable updates and it didn't give you an option to say no, it was either fully automatic or downloads only and you may have forgotten to go and disable them again in case you already had before.

This is because if you look at the release changelog there were literally no graphics related changes in 2.3.1 and 2.3.2, so if 2.3.0 works then those versions should work as well. 2.3.1 was a bug fix for vaulted OpenCore not able to boot, 2.3.2 was a fix for T1 issues in 15.4.

 

[Larsvonhier]

Make sure to disable all automatic macOS updates (all the way to downloading) as they can now modify the system volume while staging the update (which reseals the snapshot), which can lead to broken patches.

Sequoia 15.4 recently prompted to enable updates and it didn't give you an option to say no, it was either fully automatic or downloads only and you may have forgotten to go and disable them again in case you already had before.

That´s a good hint, thanks.
Only why does this not happen with earlier OCLPs?
(We never had this happen until recently with said configuration and the setting for auto-detect/download updates has not been touched).

 

[Neon Ball]

That´s a good hint, thanks.
Only why does this not happen with earlier OCLPs?
(We never had this happen until recently with said configuraion and the setting for auto-detect/download updates has not been touched).

I ninja edited to the original that 2.3.1 and 2.3.2 didn't have any graphics related changes. The update issue has been there since Sonoma, there's even a documentation part for it. It seems to be quite random when it happens.

 

[Larsvonhier]

I ninja edited to the original that 2.3.1 and 2.3.2 didn't have any graphics related changes. The update issue has been there since Sonoma, there's even a documentation part for it. It seems to be quite random when it happens.

Well, I set all related toggle switches to "off" now and restrain from running Lockrattler for security updates until really needed or just after updating macOS and before patching...
Let´s see!
Thanks.

 

[adrianlondon]

My 2024 MBP died - it's currently in the Apple Store waiting for delivery and installation of a new logic board. Whoops!

So, got my 2013 MBP - the one with the smashed screen - out of the back of my cupboard and connected it via HDMI to my monitor. It was running Big Sur.

Installed OCLP 2.3.2 and it very smoothly upgraded me to MacOS Sequoia 15.5 beta. I'm really impressed. Then I successfully restored from my latest Time Machine backup.

It's been a few years since I last used OCLP - well done to the team behind this, as it worked very well.