Kernel Panic - Yup again and again

[Washac]

Hi All

I have seen and read all over the net people with this type of panic and some have had everything in their Mac Pros changed except the case
and still they get this type of panic.

My Mac Pro started this panic after 10.6, it then as good as disappeared with each update, now it is back with vengence at almost one a day....

Always occurs when using either Firefox or Safari, others in my house use the Pro not for browsing and they never see a panic.

I have done all the Apple hardware tests I have loop tested the memory over and over and over and found no RAM problems.

Interval Since Last Panic Report: 348305 sec
Panics Since Last Report: 4
Anonymous UUID: 2E61D50D-3784-43DB-8413-946B8247D592

Wed Dec 16 18:26:40 2009
Machine-check capabilities (cpu 4) 0x0000000000001c09:
family: 6 model: 26 stepping: 5 microcode: 15
Intel(R) Xeon(R) CPU W3520 @ 2.67GHz
9 error-reporting banks
threshold-based error status present
extended corrected memory error handling present
Machine-check status 0x0000000000000004:
machine-check in progress
MCA error-reporting registers:
IA32_MC0_STATUS(0x401): 0x0000000000000800 invalid
IA32_MC1_STATUS(0x405): 0x0000000000000800 invalid
IA32_MC2_STATUS(0x409): 0x0000000000000000 invalid
IA32_MC3_STATUS(0x40d): 0x0000000000000000 invalid
IA32_MC4_STATUS(0x411): 0x0000000000000000 invalid
IA32_MC5_STATUS(0x415): 0x0000000000000000 invalid
IA32_MC6_STATUS(0x419): 0x0000000000000000 invalid
IA32_MC7_STATUS(0x41d): 0x0000000000000000 invalid
Package 0 logged:
IA32_MC8_STATUS(0x421): 0xbe0000000001009f valid
Channel number: 15 (unknown)
Memory Operation: read
Machine-specific error: Read ECC
COR_ERR_CNT: 0
Status bits:
Processor context corrupt
ADDR register valid
MISC register valid
Error enabled
Uncorrected error
IA32_MC8_ADDR(0x422): 0x0000000075b81840
IA32_MC8_MISC(0x423): 0x88707fb000081285
DIMM: 0
Channel: 2
Syndrome: 0x88707fb0
panic(cpu 4 caller 0x2a6c94): Machine Check at 0x00fb23ad, thread:0x7b0fb7c, trapno:0x12, err:0x0,registers:
CR0: 0x8001003b, CR2: 0x468b0000, CR3: 0x00100000, CR4: 0x00000660
EAX: 0x00000000, EBX: 0x61809ade, ECX: 0x00000001, EDX: 0x3d8f0210
ESP: 0x4d65bbb8, EBP: 0x4d65bbb8, ESI: 0x00001d43, EDI: 0x00000000
EFL: 0x00000046, EIP: 0x00fb23ad

Backtrace (CPU 4), Frame : Return Address (4 potential args on stack)
0x44255fb8 : 0x21b2bd (0x5cf868 0x44255fec 0x223719 0x0)
0x44256008 : 0x2a6c94 (0x5918f8 0x5919e5 0xfb23ad 0x7b0fb7c)
0x442560f8 : 0x29e8d2 (0x1 0x66654444 0x746c7561 0x61686542)
0x4d65bbb8 : 0xfb2092 (0x4417e0a4 0x80 0x54e0ee 0x54e21e)
0x4d65bbf8 : 0xfb21b6 (0x4417e0a4 0x618d878f 0x1d43 0x617de271)
0x4d65bc48 : 0xfaab4a (0x4417e0a4 0x617de271 0x80001d43 0x0)
0x4d65bd28 : 0xfac4db (0x617de271 0x80001d43 0x0 0x0)
0x4d65bdf8 : 0x2a8a3c (0xffffffff 0x7fffffff 0x5 0x4418d000)
0x4d65be18 : 0x225527 (0x4418d49c 0x617de0dd 0x1d43 0x4418d488)
0x4d65be78 : 0x2262fd (0x1 0x7b0fb7c 0x7301c58 0x11245f0)
0x4d65bee8 : 0x226f11 (0x7b0fb98 0x0 0x4d65bf48 0x54966b)
0x4d65bf58 : 0x226f92 (0x22f88a 0x863ea0 0x0 0x2a0596)
0x4d65bf78 : 0x22fa24 (0x22f88a 0x863ea0 0x0 0x0)
0x4d65bfc8 : 0x29d68c (0x863ea0 0x0 0x10 0x8c3dba4)
Kernel Extensions in backtrace (with dependencies):
com.apple.driver.AppleIntelCPUPowerManagement(96.0.0)@0xfa9000->0xfc3fff


So many out here getting this type of KP somebody has got to find or know the answer soon.

 

[lannister80]

This is the line that sticks out (for me):

Machine-specific error: Read ECC

I know memory passed all the tests, but maybe it's a weird, specific error.

Can you experiment with running the machine without one of the RAM sticks for a day or two and see if you can narrow it down to a single stick?

Are you RAM temperatures high?

Do you live at a high altitude (seriously)? Soft Error Rates (SERs) due to cosmic rays are 10x higher at 1 mile above sea level that at sea level. I'm sure SERs are still pretty rare, but maybe, just maybe...

 

[Washac]

This is the line that sticks out (for me):

Machine-specific error: Read ECC

I know memory passed all the tests, but maybe it's a weird, specific error.

Can you experiment with running the machine without one of the RAM sticks for a day or two and see if you can narrow it down to a single stick?

Are you RAM temperatures high?

Do you live at a high altitude (seriously)? Soft Error Rates (SERs) due to cosmic rays are 10x higher at 1 mile above sea level that at sea level. I'm sure SERs are still pretty rare, but maybe, just maybe...

1. Specific error, yes I have been thinking that.
2. RAM Temps 40c to 44c
3. No high Altitude, live just outside London

 

[3587]

Well, when I thought mine was the memory, they sent someone out to my house to replace all kinds of things. It wasn't until I brought it to the Apple Store did they just give me a whole new machine. They couldn't figure it out, but the memory wasn't talking to the rest of the machine correctly. So instead of replacing more hardware, they just replaced the machine. Bottom line, drop it off, it only took one day for them to figure it out and get me a whole new one. Good luck! KP's suck!

 

[Washac]

Well, when I thought mine was the memory, they sent someone out to my house to replace all kinds of things. It wasn't until I brought it to the Apple Store did they just give me a whole new machine. They couldn't figure it out, but the memory wasn't talking to the rest of the machine correctly. So instead of replacing more hardware, they just replaced the machine. Bottom line, drop it off, it only took one day for them to figure it out and get me a whole new one. Good luck! KP's suck!

Thing is how do they see the KP as it is so random and I cannot even recreate it ?

I have this idea that I call them, they take it away as I am in what they call a dead zone for call out or take in to store, then its returned with the same problem. as I have heard has happened to others

 

[3587]

Thing is how do they see the KP as it is so random and I cannot even recreate it ?

I have this idea that I call them, they take it away as I am in what they call a dead zone for call out or take in to store, then its returned with the same problem. as I have heard has happened to others

They told me they ran a program that stress tested the computer overnight... When they came in the next morning, the thing wouldn't even startup! So, they just replaced it. They have ways of making it KP. It helps to tell them exactly what was open and what you did leading up to the panic.

 

[Washac]

They told me they ran a program that stress tested the computer overnight... When they came in the next morning, the thing wouldn't even startup! So, they just replaced it. They have ways of making it KP. It helps to tell them exactly what was open and what you did leading up to the panic.

Looks like if I cannot resolve it myself, then it goes back after Christmas....

 

[3587]

Looks like if I cannot resolve it myself, then it goes back after Christmas....

Sucks, but in order to fix it for good... That is the only way.

I replaced by memory twice, new motherboard, new processor, etc. Nothingfixed it.

Get it in. Get it fixed. Re-enjoy. What a pain, I agree.

 

[lannister80]

1. Specific error, yes I have been thinking that.
2. RAM Temps 40c to 44c
3. No high Altitude, live just outside London

RAM is nice and cool. I was thinking maybe it was a rarely-addressed area of RAM that was barfing, but if you ran a bunch of memory tests and they all came back clean...

Bad RAM slot? Bad RAM riser? Obviously not fixable without going to the Apple store...

 

[Washac]

RAM is nice and cool. I was thinking maybe it was a rarely-addressed area of RAM that was barfing, but if you ran a bunch of memory tests and they all came back clean...

Bad RAM slot? Bad RAM riser? Obviously not fixable without going to the Apple store...

I live in what they call ad ead zone, so they come and pick up, I just do not want it to go
back and forth untill they decide to give a new machine....

I did drop into the System Profiler once and I see some memory correctable errors.

 

[lannister80]

I live in what they call ad ead zone, so they come and pick up, I just do not want it to go
back and forth untill they decide to give a new machine....

I did drop into the System Profiler once and I see some memory correctable errors.

Do they show up on just one RAM module, or on random ones? If you can narrow it down to a module, remove/RMA it.

 

[Washac]

Do they show up on just one RAM module, or on random ones? If you can narrow it down to a module, remove/RMA it.

KP always shows DIMM 0 Channel 2 which is Slot 3.

Going to remove stick in that slot this weekend and see how I go, then switch that stick into slots 1 and 2 to see if the problem moves.

RMA it ?

 

[nanofrog]

Have you tried pulling the riser and the memory from the riser, and cleaning all the contacts?

It could be that simple, and is easy enough to try. Seriously. Just use 91% Isopropyl alcohol (or better, if available in the UK), and a lint free cloth (i.e. the nylon handkerchiefs sold for cleaning eye glasses will work, as will ready-to-use wipes from Tech Spray).

 

[Washac]

Have you tried pulling the riser and the memory from the riser, and cleaning all the contacts?

It could be that simple, and is easy enough to try. Seriously. Just use 91% Isopropyl alcohol (or better, if available in the UK), and a lint free cloth (i.e. the nylon handkerchiefs sold for cleaning eye glasses will work, as will ready-to-use wipes from Tech Spray).

Thanks will try that also

Hang on remove riser from the riser ?

Will get some contact cleaner or wipes....

 

[nanofrog]

Hang on remove riser from the riser ?

From the logic board.

Pull a riser, and clean the contacts (where it plugs into the logic board). Then each stick stuffed in each riser. Repeat with the the second riser, and each stick in it.

 

[Washac]

From the logic board.

Pull a riser, and clean the contacts (where it plugs into the logic board). Then each stick stuffed in each riser. Repeat with the the second riser, and each stick in it.

Sorry did not realise the risers could be removed from the board

 

[nanofrog]

Sorry did not realise the risers could be removed from the board

I should have clarified. The risers are removable in the '06 - '08 systems from the logic board. Not quite the case in the '09's, as you'd need to pull the entire daughterboard, as the DIMM slots are soldered directly to it.

It would be much easier to clean each DIMM though, and try that first, on the off chance it will work.

Also, if you've not tried it, do a full shut down, wait a few minutes, then restart after cleaning the DIMM's in each slot (make sure the alcohol is completely evaporated prior to re-installing the DIMM's first). Worth a shot, at least, as it's really inexpensive (alcohol or ready-to-go wipes), and not that hard to do.

Again, sorry about the confusion.

 

[Washac]

I should have clarified. The risers are removable in the '06 - '08 systems from the logic board. Not quite the case in the '09's, as you'd need to pull the entire daughterboard, as the DIMM slots are soldered directly to it.

It would be much easier to clean each DIMM though, and try that first, on the off chance it will work.

Also, if you've not tried it, do a full shut down, wait a few minutes, then restart after cleaning the DIMM's in each slot (make sure the alcohol is completely evaporated prior to re-installing the DIMM's first). Worth a shot, at least, as it's really inexpensive (alcohol or ready-to-go wipes), and not that hard to do.

Again, sorry about the confusion.

No problem, thanks...