Help Deciphering MCA Error Report

[MacPhotog]

Hi all,

I have a "New" Mac Pro (6,1) for work, and it has started locking randomly; the machine freezes completely (clock stops, mouse & keyboard unresponsive, force-quit commands don't work, etc.) and does so regardless of applications that may be running (it will lock even when only the Finder is running).

The machine appears to realize it is locked though, because after 30 seconds or a minute or so it automatically reboots itself. After one of these instances the machine offered me an error report, and I coped the MCA Error Report.

I'm hoping some of you folks might be able to read it for me and tell me what it means. It **appears** to me to be indicating a hardware problem (and that seems reasonable given the machine's behavior).

Anyway, here goes. Any help deciphering what this means would be greatly appreciated!

*** MCA Error Report ***
CPU Machine Check Architecture Error Dump (CPU: Intel(R) Xeon(R) CPU E5-1680 v2 @ 3.00GHz, CPUID: 0x306E4)
Core: 0 1
IA32_MC3_STATUS=0xF200000000800400
IA32_MC3_CTL=0x0
Core: 0
IA32_MC21_STATUS=0xFE200000000C110A
IA32_MC21_CTL=0x0
IA32_MC21_ADDR=0xA0B00080
IA32_MC21_MISC=0x88FC381602802086
Core: 0
IA32_MC23_STATUS=0xFE200000000C110A
IA32_MC23_CTL=0x0
IA32_MC23_ADDR=0xB5E02040
IA32_MC23_MISC=0x58FC381610802086


Model: MacPro6,1, BootROM MP61.0116.B11, 8 processors, 8-Core Intel Xeon E5, 3 GHz, 32 GB, SMC 2.20f18
Graphics: AMD FirePro D700, AMD FirePro D700, PCIe, 6144 MB
Graphics: AMD FirePro D700, AMD FirePro D700, PCIe, 6144 MB
Memory Module: DIMM1, 8 GB, DDR3 ECC, 1866 MHz, 0x802C, 0x31384A534631473732415A2D314739453220
Memory Module: DIMM2, 8 GB, DDR3 ECC, 1866 MHz, 0x802C, 0x31384A534631473732415A2D314739453220
Memory Module: DIMM3, 8 GB, DDR3 ECC, 1866 MHz, 0x802C, 0x31384A534631473732415A2D314739453220
Memory Module: DIMM4, 8 GB, DDR3 ECC, 1866 MHz, 0x802C, 0x31384A534631473732415A2D314739453220
AirPort: spairport_wireless_card_type_airport_extreme (0x14E4, 0x135), Broadcom BCM43xx 1.0 (7.15.159.13.12)
Bluetooth: Version 4.3.2f6 15235, 3 services, 27 devices, 1 incoming serial ports
Network Service: Ethernet 1, Ethernet, en0
PCI Card: AMD FirePro D700, Display Controller, Slot-1
PCI Card: AMD FirePro D700, Display Controller, Slot-2
PCI Card: pci1b21,612, AHCI Controller, Thunderbolt@158,0,0
PCI Card: Apple 57761-B0, Ethernet Controller, Thunderbolt@234,0,0
PCI Card: pci11c1,5901, IEEE 1394 Open HCI, Thunderbolt@233,0,0
PCI Card: pci12d8,400e, USB Open Host Controller, Thunderbolt@239,0,0
PCI Card: pci12d8,400e, USB Open Host Controller, Thunderbolt@239,0,1
PCI Card: pci12d8,400f, USB Enhanced Host Controller, Thunderbolt@239,0,2
Serial ATA Device: APPLE SSD SM0256F, 251 GB
Serial ATA Device: Elite Pro Dual TBU3 - RAID 1, 3 TB
USB Device: USB3.0 Hub
USB Device: USB3.0 Hub
USB Device: USB_3_0 Reader
USB Device: Hub
USB Device: BRCM20702 Hub
USB Device: Bluetooth USB Host Controller
USB Device: USB2.0 Hub
USB Device: USB2.0 Hub
USB Device: Optical Drive
USB Device: Hub
USB Device: Keyboard Hub
USB Device: Apple Keyboard
USB Device: FaceTime HD Camera (Display)
USB Device: Apple Thunderbolt Display
USB Device: Display Audio
Thunderbolt Bus: Mac Pro, Apple Inc., 19.2
Thunderbolt Bus: Mac Pro, Apple Inc., 19.2
Thunderbolt Device: Elite Pro Dual TBU3, Other World Computing, 1, 30.2
Thunderbolt Bus: Mac Pro, Apple Inc., 19.2
Thunderbolt Device: Thunderbolt Display, Apple Inc., 3, 26.2

 

[Machines]

This is a MCE (Machine Check Exception) . Your Mac's CPU says your system most likely has a hardware error , not otherwise identifiable .

To verify it is not a software error , do you have access to an external USB drive with just a clean install of Mavericks OS X 10.9.5 ? I would not recommend Yosemite . Don't have any third party apps installed on this test drive since we want to make certain an app or bad OS element (i.e. kext) is not causing the error . Connect it and boot from this drive .

If you don't , some possible reasons for the error are faulty ECC memory modules and an overloaded PSU (i.e. you are stressing out your Mac .)

Please go into system profile , to where memory is listed and see if any modules are showing ECC errors . If not, run Memtest for Mac and see if the ram has passed diagnostics .

There isn't really much else you can do , except these three things, since Apple and Intel do not share error codes .

 

[goMac]

If you don't , some possible reasons for the error are faulty ECC memory modules and an overloaded PSU (i.e. you are stressing out your Mac .)

Keep in mind, you shouldn't be able to crash your Mac Pro by stressing it out. So I'd consider this a hardware fault any way you look at it. I'd contact Apple support.

 

[MacPhotog]

Eesh.

I discovered (completely by accident) that the issue that was causing the crashes was actually an external RAID unit with a faulty SATA cable:

I have an OWC Mercury Elite Pro Dual RAID1 box, and months ago OWC recognized a flaw in the original cables and sent a replacement (saying that the faulty cable could cause the unit to drop offline). Mine wasn't acting up so I didn't replace the cable.

Lo and behold this morning my entire system starts crashing. Because I have a ton of important data on that RAID unit I disconnected it to be safe, and all of a sudden the crashes stopped. I attached the RAID unit to a laptop (which had never had any issues) and all of a sudden the laptop started crashing. Bingo.

Anyway, I've now replaced the cable and the crashes have stopped and the system is functioning normally.

But that still leaves the issue of the error codes I was receiving in the MCA report.... some folks responded that it looked like it could be a RAM issue, etc... I don't currently have the tools on this machine to do the necessary diagnostics on the RAM, etc.

Do we think it was just the RAID unit and I can go on blissfully in the belief that nothing is wrong with the machine itself? Or do those error codes indicate something was also, simultaneously wrong with the machine that I need to get looked at before the machine's warranty runs up?

Many thanks for all your help!

 

[goMac]

A Thunderbolt array could probably cause those errors. Hopefully that was it!

 

[Machines]

Keep in mind, you shouldn't be able to crash your Mac Pro by stressing it out. So I'd consider this a hardware fault any way you look at it. I'd contact Apple support.

I wish that was true . But, one of the reasons why I perform long burn ins with my Mac Pro builds is to determine if they crash under heavy load (multiple component subsystems each at load .)

----------

Eesh.

I discovered (completely by accident) that the issue that was causing the crashes was actually an external RAID unit with a faulty SATA cable:

I have an OWC Mercury Elite Pro Dual RAID1 box, and months ago OWC recognized a flaw in the original cables and sent a replacement (saying that the faulty cable could cause the unit to drop offline). Mine wasn't acting up so I didn't replace the cable.

Lo and behold this morning my entire system starts crashing. Because I have a ton of important data on that RAID unit I disconnected it to be safe, and all of a sudden the crashes stopped. I attached the RAID unit to a laptop (which had never had any issues) and all of a sudden the laptop started crashing. Bingo.

Anyway, I've now replaced the cable and the crashes have stopped and the system is functioning normally.

But that still leaves the issue of the error codes I was receiving in the MCA report.... some folks responded that it looked like it could be a RAM issue, etc... I don't currently have the tools on this machine to do the necessary diagnostics on the RAM, etc.

Do we think it was just the RAID unit and I can go on blissfully in the belief that nothing is wrong with the machine itself? Or do those error codes indicate something was also, simultaneously wrong with the machine that I need to get looked at before the machine's warranty runs up?

Many thanks for all your help!

If your Mac is stable , then problem solved . You're looking for an unimportant needle in an unnecessary haystack .

But if you want to feel better, grab a copy of Memtest for Mac and just wait for the test to pass . Enjoy your several pots of coffee , because it's a long test .

 

[Anim]

On my 6,1 I have been having this problem too. I am sure it is my external Lacie 4TB Thunderbolt. Mine locks up in Windows Bootcamp though. Machine freezes for 30s - 2 mins then lets me move the mouse for about 10 seconds. To fix it all, I do a single reboot and the freezes are gone for the next session. When I get back into macOS I get the system crash report so it must get it from bios or whatever the mac uses.

A new TB cable from Lacie has not solved this. The Lacie is on its own thunderbolt bus.

Tue Dec 6 11:59:08 2016

*** MCA Error Report ***
CPU Machine Check Architecture Error Dump (CPU: Intel(R) Xeon(R) CPU E5-1650 v2 @ 3.50GHz, CPUID: 0x306E4)
Core: 0
IA32_MC18_STATUS=0xFE200000000C110A
IA32_MC18_CTL=0x0
IA32_MC18_ADDR=0x8F3A0400
IA32_MC18_MISC=0x5CFC385604802086
Core: 4 5
IA32_MC3_STATUS=0xF200000000800400
IA32_MC3_CTL=0x0


Model: MacPro6,1, BootROM MP61.0116.B20, 6 processors, 6-Core Intel Xeon E5, 3.5 GHz, 16 GB, SMC 2.20f18
Graphics: AMD FirePro D700, AMD FirePro D700, PCIe, 6144 MB
Graphics: AMD FirePro D700, AMD FirePro D700, PCIe, 6144 MB
Memory Module: DIMM1, 4 GB, DDR3 ECC, 1866 MHz, 0x02FE, 0x45424A3430454738424657422D4A532D4620
Memory Module: DIMM2, 4 GB, DDR3 ECC, 1866 MHz, 0x02FE, 0x45424A3430454738424657422D4A532D4620
Memory Module: DIMM3, 4 GB, DDR3 ECC, 1866 MHz, 0x02FE, 0x45424A3430454738424657422D4A532D4620
Memory Module: DIMM4, 4 GB, DDR3 ECC, 1866 MHz, 0x02FE, 0x45424A3430454738424657422D4A532D4620

 

[Chasesurf91]

A Thunderbolt array could probably cause those errors. Hopefully that was it!

Wow I'm so glad I found this discussion. Just got the same error on my 2016 mbp. I spent the better part of a week trying to rid my computer of the hackers who must have been doing this to my little computer.

I too have a Lacie external hd. 5tb p2337 and oddly enough I did something to the USB c connector and its bent at like 20 Degrees still registered when I shoved it back it in so I figured it must be just fine.

I'm still within the return window for the hd should I take this thing back? It's so pretty and shiny and well sealed though and my last Porsche drive is like 20 years old and aging gracefully. Although it has been making an angry sound at shut off and I can't tell if it's just old age or jealousy. I feel like Seagate's (lacie) quality went down a little since I bought my last one.

My memory is showing as ECC: Disabled is that bad or good?