MP 1,1-5,1 Mojave random kernel panic on Mac Pro 4.1 -> 5.1

[jscipione]

After you do the northbridge maintenance, check if you have a healthy NVRAM volume. The KPs are saved inside the NVRAM stores and make a real mess that overtime can become a corrupt volume/brick.

OpenCore on the Mac Pro

Thank you @tsialex as always! Your knowledge overwhelms me! I have fixed my problem just reflashing BootROM using the reconstructed one that you made it for me. Now everything is ok again

forums.macrumors.com

Not to hijack thread but... I also did a northbridge repaste due to restarts and I also have a cross-flashed 4,1 => 5,1 dual socket Mac Pro. I dumped my rom but I don't seem to have a Free space variable under VSS store. Can you tell me what I'm doing wrong?

 

[tsialex]

Not to hijack thread but... I also did a northbridge repaste due to restarts and I also have a cross-flashed 4,1 => 5,1 dual socket Mac Pro. I dumped my rom but I don't seem to have a Free space variable under VSS store. Can you tell me what I'm doing wrong?

That free space below the Gaid store is normal and expected, it's the free space for the full NVRAM volume. You need to look for the free space inside the VSS stores.

NVRAM main VSS store is already corrupt - correction, the screenshot show that both stores are corrupt. The padding area is where the corrupt mumbo-jumbo of variables are.

No free space being shown inside the VSS stores, even if it's one byte, is a confirmation that the VSS store/stores is/are now corrupt.

Check your PMs.

 

[Aidzer0]

Just as a side note, follow the DM instructions, it’s very informative and very helpful.

I don’t mind thread jacking at all, more info that’s spread, the more it will help others

I’m Very glad to have a community like this with people willing to help.

 

[Aidzer0]

So far so good, I haven’t had a KP after reposting and changing out those push pins for the north bridge, however I have been noticing that the north ridge diode gets to about 70°C which I’m still not too comfortable with.

I found a really good thread experimenting with diverting airflow to focus a little more on the NB heatsink. So naturally I will be attempting that.

Thanks all for the awesome guide and helping me resolve this issue!

 

[KeesMacPro]

So far so good, I haven’t had a KP after reposting and changing out those push pins for the north bridge, however I have been noticing that the north ridge diode gets to about 70°C which I’m still not too comfortable with.

I found a really good thread experimenting with diverting airflow to focus a little more on the NB heatsink. So naturally I will be attempting that.

Thanks all for the awesome guide and helping me resolve this issue!

If the MP is dustfree , IMHO there is a very effective and easy way to keep the NB Temp under control.
Supposing MacsFanControl or similar is installed you could set only 2 fans like this:
- Boost A fan: min 55 C max 75 C sensor: NB diode
- Intake : min 55 C max 75 C sensor: NB diode

With all other fans set to auto, your Mac Pro will stay well within a safe Temp range regardless the load.

 

[Aidzer0]

If the MP is dustfree , IMHO there is a very effective and easy way to keep the NB Temp under control.
Supposing MacsFanControl or similar is installed you could set only 2 fans like this:
- Boost A fan: min 55 C max 75 C sensor: NB diode
- Intake : min 55 C max 75 C sensor: NB diode

With all other fans set to auto, your Mac Pro will stay well within a safe Temp range regardless the load.

Interesting, I will try that first, thanks!

 

[Aidzer0]

If the MP is dustfree , IMHO there is a very effective and easy way to keep the NB Temp under control.
Supposing MacsFanControl or similar is installed you could set only 2 fans like this:
- Boost A fan: min 55 C max 75 C sensor: NB diode
- Intake : min 55 C max 75 C sensor: NB diode

With all other fans set to auto, your Mac Pro will stay well within a safe Temp range regardless the load.

30 minutes of load and did not spike past 62°C, much better than 72 - 74°C, really good advice, thank you! No need for any mods.

 

[Aidzer0]

Just a quick update on everything, went ahead with tsialex's reconstruction, holy moly, very worth it.

Also, I did run ASD, all passed without problems.

Much snappier and most of the issues I had are gone with the fresh BootROM flashed.

The re-paste and push pin/clip replacement worked out, also changing the fan speeds on Boost A fan and Intake got the NB diode down to 60°C constant now. My ambient is 30°C (Actually 23°C but Macs Fan Controller states 30°C). So all in all, seems that everything went smoothly.

I currently am having an issue with USB HDD, it's a 3.0 USB External (2.5") which seems to disconnect after being connected for about 10 minutes. That being said, I am using the stock 2.0 USB ports, which are probably not giving enough power to the USB 3.0 drive. That I need to research more.

Thanks all to the input and help!

 

[tsialex]

Just a quick update on everything, went ahead with tsialex's reconstruction, holy moly, very worth it.

Also, I did run ASD, all passed without problems.

Much snappier and most of the issues I had are gone with the fresh BootROM flashed.

The re-paste and push pin/clip replacement worked out, also changing the fan speeds on Boost A fan and Intake got the NB diode down to 60°C constant now. My ambient is 30°C (Actually 23°C but Macs Fan Controller states 30°C). So all in all, seems that everything went smoothly.

Glad that everything worked out.

I currently am having an issue with USB HDD, it's a 3.0 USB External (2.5") which seems to disconnect after being connected for about 10 minutes. That being said, I am using the stock 2.0 USB ports, which are probably not giving enough power to the USB 3.0 drive. That I need to research more.

Thanks all to the input and help!

Mac Pro native USB ports are on the weak side energy wise, AFAIK it's just 500+500mA. Lot's of power hungry bus powered drives have problems with it.

 

[Aidzer0]

Glad that everything worked out.


Mac Pro native USB ports are on the weak side energy wise, AFAIK it's just 500+500mA. Lot's of power hungry bus powered drives have problems with it.

Yeah that’s what I figured, I’ll just get a usb 3.1 PCI card to resolve that issue.

Thanks again.

 

[Ludor]

Jiminy. I wonder if I am having the same problem. I have an 4,1>5,1 with dual 5690, and it keeps shutting down about once a day. I have never heard about the nvram stores, but it sounds like I have to look into it.

 

[Aidzer0]

Jiminy. I wonder if I am having the same problem. I have an 4,1>5,1 with dual 5690, and it keeps shutting down about once a day. I have never heard about the nvram stores, but it sounds like I have to look into it.

Make sure to check the North bridge, repaste and replace those clips if you haven’t already.

That’s the main cause for the KP.

 

[tsialex]

Make sure to check the North bridge, repaste and replace those clips if you haven’t already.

That’s the main cause for the KP.

You are correct, but once you have frequent KPs the damage inside the NVRAM volume/BootROM is done and even if you correctly repaired the northbridge the mess inside the NVRAM still cause adverse effects.

 

[Ludor]

You are correct, but once you have frequent KPs the damage inside the NVRAM volume/BootROM is done and even if you correctly repaired the northbridge the mess inside the NVRAM still cause adverse effects.

Well, I know very little but I know not to question you about Macs. It’s just that I’ve never heard about this issue, kernel panics clogging up the NVRAM, before. But as you pointed out, these are 12 year old computers. I managed to find ROMtool and will try it tomorrow and see what kind of results I get. And just resetting NVRAM won’t help anything?

 

[tsialex]

Well, I know very little but I know not to question you about Macs. It’s just that I’ve never heard about this issue, kernel panics clogging up the NVRAM, before. But as you pointed out, these are 12 year old computers. I managed to find ROMtool and will try it tomorrow and see what kind of results I get. And just resetting NVRAM won’t help anything?

This issue is a lot more complex and nuanced. There are lot's of classes of data stored inside the NVRAM. Some are public, some private, some user accessible, some not. The way the data is stored is counter-intuitive also, it's a circular log similar to JFFS2.

A NVRAM reset really only clears the data that are public and user accessible. If the NVRAM reset worked as it's usually implied, like the older PRAM worked with OpenFirmare Macs, the serial numbers/hardwareIDs or MDM/management data wouldn't be permanent.

Another issue, for the NVRAM reset procedure to work, the garbage collection needs to be fully working, if it's not, nothing is cleared inside the NVRAM stores.

 

[Ludor]

This issue is a lot more complex and nuanced. There are lot's of classes of data stored inside the NVRAM. Some are public, some private, some user accessible, some not. The way the data is stored is counter-intuitive also, it's a circular log similar to JFFS2.

A NVRAM reset really only clears the data that are public and user accessible. If the NVRAM reset worked as it's usually implied, like the older PRAM worked with OpenFirmare Macs, the serial numbers/hardwareIDs or MDM/management data wouldn't be permanent.

Another issue, for the NVRAM reset procedure to work, the garbage collection needs to be fully working, if it's not, nothing is cleared inside the NVRAM stores.

Thank you kindly. I will try to follow the guide in the OpenCore thread for how to deal with it. But first I must find all the material for taking care of the NorthBridge.

 

[tsialex]

Thank you kindly. I will try to follow the guide in the OpenCore thread for how to deal with it. But first I must find all the material for taking care of the NorthBridge.

Check the northbridge push pins first, then follow the instructions on the PM I've sent you - I'll check your BootROM image.

 

[Ludor]

Check the northbridge push pins first, then follow the instructions on the PM I've sent you - I'll check your BootROM image.

Thank you, and sorry for not responding for over a month but I could not find the time in between work to take my Mac apart.

Now I have done that… repeatedly… and after many attempts (blinking power LED, dark efi_done LED – it somehow came back, though – I got the poor thing to boot from my old Lion HDD. The machine still won’t boot from my main startup disk, an SSD in the disc burner compartment.

So, one question: Could all the kernel panics clogging up the BootROM prevent a proper boot procedure?

 

[tsialex]

Thank you, and sorry for not responding for over a month but I could not find the time in between work to take my Mac apart.

Now I have done that… repeatedly… and after many attempts (blinking power LED, dark efi_done LED – it somehow came back, though – I got the poor thing to boot from my old Lion HDD. The machine still won’t boot from my main startup disk, an SSD in the disc burner compartment.

So, one question: Could all the kernel panics clogging up the BootROM prevent a proper boot procedure?

Yes and this issue can evolve to a brick easily. You'll probably need a BootROM reconstruction or a reconstruction and SPI flash replacement.

 

[Ludor]

Yes and this issue can evolve to a brick easily. You'll probably need a BootROM reconstruction or a reconstruction and SPI flash replacement.

Yes, I see. But how would one do that? Can RomTool be run from my Lion disk, or the Recovery partition?

I better go read that kind DM you sent last month again.

Edit: The reason I ask is because I don’t know from where to run RomTool, in order to dump my rom and send it for replacement.

 

[tsialex]

Yes, I see. But how would one do that? Can RomTool be run from my Lion disk, or the Recovery partition?

I better go read that kind DM you sent last month again.

Edit: The reason I ask is because I don’t know from where to run RomTool, in order to dump my rom and send it for replacement.

Lion won't work. 10.9 works, 10.10 don't, 10.11 to 10.14.

 

[Ludor]

(with apologies if I am repeating myself) So I have a 4,1>5,1 with Mojave on an SSD and a secondary startup Lion HDD. I’ve been having spontaneous reboots, and I finally took my machine apart and cleaned and repasted the NorthBridge (and put in new bolts).

After that, I have never got back into Mojave. I learned about the diagnostic LEDs and it seemed my backplane was gone, but then its LED lighted up again. Yesterday I could boot Lion (although suspiciously slowly, like the keyboard didn’t work at the login for five minutes).

I still haven’t been able to upgrade Lion to an OS that is compatible with ROMTool.

Some attempts to boot into Lion on a HDD, without the troublesome SSD plugged in, have worked. For a while there, I had a green EFI_DONE led back, but it’s dark again. On some boots, the GPU led was blinking, but it’s green now. Fans all go at a steady rate.

Where I am now, I get a truncated chime and a grey screen (no apple).

I really don’t know what to try at this point. Any suggestions are very welcome.

 

[tsialex]

(with apologies if I am repeating myself) So I have a 4,1>5,1 with Mojave on an SSD and a secondary startup Lion HDD. I’ve been having spontaneous reboots, and I finally took my machine apart and cleaned and repasted the NorthBridge (and put in new bolts).

After that, I have never got back into Mojave. I learned about the diagnostic LEDs and it seemed my backplane was gone, but then its LED lighted up again. Yesterday I could boot Lion (although suspiciously slowly, like the keyboard didn’t work at the login for five minutes).

I still haven’t been able to upgrade Lion to an OS that is compatible with ROMTool.

Some attempts to boot into Lion on a HDD, without the troublesome SSD plugged in, have worked. For a while there, I had a green EFI_DONE led back, but it’s dark again. On some boots, the GPU led was blinking, but it’s green now. Fans all go at a steady rate.

Where I am now, I get a truncated chime and a grey screen (no apple).

I really don’t know what to try at this point. Any suggestions are very welcome.

Seems you have more than one problem at the same time, this will require testing the components of the defective Mac Pro with a known working one, to eliminate what is not working correctly.

I'd start testing the PSU with a known working Mac Pro and installing the known working PSU on the defective one. After eliminating the PSU, test the CPU tray and GPU. Test just one component at a time.

 

[Ludor]

Seems you have more than one problem at the same time, this will need to test the components with a known working Mac Pro.

I'd start testing the PSU with a known working Mac Pro and installing the known working PSU on the defective one. After eliminating the PSU, test the CPU tray.

Thank you, sir. Do you happen to know if the PSU in a 2008 Mac Pro would be equivalent?

 

[tsialex]

Thank you, sir. Do you happen to know if the PSU in a 2008 Mac Pro would be equivalent?

Not compatible, totally different connector and internally.