2009 Mac Pro frequent kernel panics - RESOLVED
Posted: Sun Jul 23, 2017 9:40 pm
I've been getting frequent kernel panics the past few days on my 2009 Mac Pro 8 x 2.26 GHz "Nehalem".
I've done a bunch of troubleshooting, but am not yet sure if it's a hardware or software problem. It's happened while working in several different apps - nothing consistent.
Steps taken, not in chronological order: reinstalled El Capitan from Recovery Mode, used Disk Utility to perform first aid on System drive while in Recovery Mode, reset NVRAM, reset SMC, used Tech Tool Pro to repair permissions and check the RAM, disconnected all peripherals except the display and a USB hub for my keyboard and trackpad, checked the system drive with Onyx, booted in Safe Mode, removed all RAM except the original RAM, removed the original RAM and left in the newer RAM.
Based on some stuff I read online, I checked the panic logs, especially the section "Kernel Extensions in backtrace".
Some of them mention "com.apple.driver.AppleIntelCPUPowerManagement".
Others mention what I think has to do with the graphics card, "com.apple.nvidia.classic.NVDAResmanTesla". Except from what I've read online, a bad graphics card would probably show symptoms like a gray or black screen.
As I write this, I've been booted in Safe Mode for almost two hours, and haven't yet had a crash. So maybe it's a software issue.
The panics started soon after I installed the latest security update for El Capitan (along with a few other app updates, including Logic Pro, MainStage, and Postbox).
Anybody had a similar experience and figured it out?
EDIT: Here's the solution, to save you reading the whole thread. In between CPU A and CPU B is a chip called the Northbridge (which controls traffic to the CPUs). It has its own little heat sink with vertical metal fins. One of the two plastic push-pins which secures the heat sink to the chip had disintegrated. The Northbridge chip was overheating, causing the panics and shutdowns.
See more here: http://www.xlr8yourmac.com/archives/jul ... verheating
The heat sink for CPU A partially covers the Northbridge heat sink, including one of the push pins. So you need to remove the heat sink (and the processor, which will probably stick to the bottom of the heat sink) in order to check or replace it.
To remove the CPU heat sink, you need a long 3mm hex wrench to reach the captive bolts, like the Eklind 54930.
https://www.amazon.com/Eklind-54930-Cus ... B000X285AW
Push-pins are hard to find locally, and ones from eBay have to be altered to be used, but as mentioned in a Mac Rumors post, I used #4-40 machine screws with nuts.
Hope this helps.
Bob
I've done a bunch of troubleshooting, but am not yet sure if it's a hardware or software problem. It's happened while working in several different apps - nothing consistent.
Steps taken, not in chronological order: reinstalled El Capitan from Recovery Mode, used Disk Utility to perform first aid on System drive while in Recovery Mode, reset NVRAM, reset SMC, used Tech Tool Pro to repair permissions and check the RAM, disconnected all peripherals except the display and a USB hub for my keyboard and trackpad, checked the system drive with Onyx, booted in Safe Mode, removed all RAM except the original RAM, removed the original RAM and left in the newer RAM.
Based on some stuff I read online, I checked the panic logs, especially the section "Kernel Extensions in backtrace".
Some of them mention "com.apple.driver.AppleIntelCPUPowerManagement".
Others mention what I think has to do with the graphics card, "com.apple.nvidia.classic.NVDAResmanTesla". Except from what I've read online, a bad graphics card would probably show symptoms like a gray or black screen.
As I write this, I've been booted in Safe Mode for almost two hours, and haven't yet had a crash. So maybe it's a software issue.
The panics started soon after I installed the latest security update for El Capitan (along with a few other app updates, including Logic Pro, MainStage, and Postbox).
Anybody had a similar experience and figured it out?
EDIT: Here's the solution, to save you reading the whole thread. In between CPU A and CPU B is a chip called the Northbridge (which controls traffic to the CPUs). It has its own little heat sink with vertical metal fins. One of the two plastic push-pins which secures the heat sink to the chip had disintegrated. The Northbridge chip was overheating, causing the panics and shutdowns.
See more here: http://www.xlr8yourmac.com/archives/jul ... verheating
The heat sink for CPU A partially covers the Northbridge heat sink, including one of the push pins. So you need to remove the heat sink (and the processor, which will probably stick to the bottom of the heat sink) in order to check or replace it.
To remove the CPU heat sink, you need a long 3mm hex wrench to reach the captive bolts, like the Eklind 54930.
https://www.amazon.com/Eklind-54930-Cus ... B000X285AW
Push-pins are hard to find locally, and ones from eBay have to be altered to be used, but as mentioned in a Mac Rumors post, I used #4-40 machine screws with nuts.
Hope this helps.
Bob