PDA

View Full Version : Dual 1.25 G4 Kernel Panics




shaneb
Jun 24, 2008, 10:29 PM
Hey there guys, I have been having a good number of kernel panics on my favorite mac box lately (last two weeks). Wonder if anyone here can decode the greek and help me out with this log.

Below is the latest log, but first a bit of history. I have run 10.5 on this machine for sometime now with very little problems. But lately I have had some massive kernel panics. I ran the Apple Hardware Test, and got a memory error, sorted through the 4 chips until I found the culprit, removed it and then the computer passed all the checks in AHT 2.0.

Right after that I booted and decided to try a little stress testing by playing back several videos with Quicktime. everything was smooth for nearly 30 minutes when I got a sudden kernel panic (the one below) The CPU at the time was running 134 degrees F, which is typical for this machine.

ANY help would be outstanding. The machine is a FW800 Dual 1.25G4 2gigs (1.5 without the "bad" chip) 160gig drive. Usually has usb 2.0 and SCSI, but the cards were removed when I started having problems.

Thanks in Advance!
ShaneB

Tue Jun 24 21:48:22 2008
Unresolved kernel trap(cpu 0): 0x400 - Inst access DAR=0x00000000240A0A08 PC=0x00000000FF8A2E60
Latest crash info for cpu 0:
Exception state (sv=0x3c362780)
PC=0xFF8A2E60; MSR=0x40001030; DAR=0x240A0A08; DSISR=0x40000000; LR=0x000AED88; R1=0x3C35FE40; XCP=0x00000010 (0x400 - Inst access)
Backtrace:
0x000395E0 0x000B05D4
Proceeding back via exception chain:
Exception state (sv=0x3c362780)
previously dumped as "Latest" state. skipping...
Exception state (sv=0x45d47500)
PC=0x9344A4B8; MSR=0x0000F030; DAR=0x240A0A08; DSISR=0x0A000000; LR=0x9348D900; R1=0xF2EB6C80; XCP=0x00000030 (0xC00 - System call)

BSD process name corresponding to current thread: QuickTime Player

Mac OS version:
9D34

Kernel version:
Darwin Kernel Version 9.3.0: Fri May 23 00:51:20 PDT 2008; root:xnu-1228.5.18~1/RELEASE_PPC
System model name: PowerMac3,6
panic(cpu 0 caller 0xFFFF0004): 0x400 - Inst access
Latest stack backtrace for cpu 0:
Backtrace:
0x0009B498 0x0009BE3C 0x00029DD8 0x000AF390 0x000B2A78
Proceeding back via exception chain:
Exception state (sv=0x3c362780)
PC=0xFF8A2E60; MSR=0x40001030; DAR=0x240A0A08; DSISR=0x40000000; LR=0x000AED88; R1=0x3C35FE40; XCP=0x00000010 (0x400 - Inst access)
Backtrace:
0x000395E0 0x000B05D4
Exception state (sv=0x45d47500)
PC=0x9344A4B8; MSR=0x0000F030; DAR=0x240A0A08; DSISR=0x0A000000; LR=0x9348D900; R1=0xF2EB6C80; XCP=0x00000030 (0xC00 - System call)

BSD process name corresponding to current thread: QuickTime Player



Firefly2002
Jun 24, 2008, 11:00 PM
I don't know what any of that means other than it's saying QT is to blame, but you might try running DiskWarrior or TechTool Pro.

Also, I don't think you know the temp of your G4 unless you're physically monitoring it with an external thermal monitor, since the built-in thermal monitors in the G4s are completely unreliable..... at least the 7400/7410s were. I'm pretty sure they didnt' fix it.

shaneb
Jun 24, 2008, 11:32 PM
I don't know what any of that means other than it's saying QT is to blame, but you might try running DiskWarrior or TechTool Pro.

Also, I don't think you know the temp of your G4 unless you're physically monitoring it with an external thermal monitor, since the built-in thermal monitors in the G4s are completely unreliable..... at least the 7400/7410s were. I'm pretty sure they didnt' fix it.

Thanks for the reply. The FW800 (&MDD) machines have a dedicated temperature monitor like the G5s. And the temps are in line with what is normal, so I think I am ok there.

I have run disk utility and diskwarrior on this drive (just a few weeks old and a fresh install of 10.5) and neither report any problems.

Firefly2002
Jun 24, 2008, 11:54 PM
I don't know what you mean by "dedicated thermal monitor".... you mean one not on the CPU but inside the G4 case?

Have you tried a CUDA reset? NVRAM? PRAM?

shaneb
Jun 25, 2008, 07:58 AM
I don't know what you mean by "dedicated thermal monitor".... you mean one not on the CPU but inside the G4 case?

Have you tried a CUDA reset? NVRAM? PRAM?

Nah, it's on the CPU, anyhow, I am convinced that heat is not the problem here.

And yeah, all the above have been reset several times.

Tracer
Jun 25, 2008, 08:25 AM
Try running the memory test with 1 Ram stick at a time to figure out if any of the RAM is bad.

Tracer

shaneb
Jun 25, 2008, 08:38 AM
Try running the memory test with 1 Ram stick at a time to figure out if any of the RAM is bad.

Tracer

I already did, but let me explain what I did. I ran Rember with all the ram in, and there were no errors.

BUT, when I tried to boot AHT, I got a memory error. So then I sorted through the chips one at a time and then removed the culprit.

So maybe there is something in that (OEM) stick that the computer doesn't like, but doesn't show up in Rember?

shaneb
Jun 27, 2008, 05:45 PM
New crash... Went nearly three days without a crash, no problems. Then out of the blue while iTunes was streaming radio (nothing else happening, I was across the room) boom... PANIC! Looks like it mentions the ATI Radeon 8500 driver... funny thing... I have an ATI Radeon 9000 pro?

Any help appreciated. I can handle one crash every two or three days, but I still want to know what it is.




Fri Jun 27 17:38:20 2008


Unresolved kernel trap(cpu 0): 0x300 - Data access DAR=0x0000000000000010 PC=0x000000000006E600
Latest crash info for cpu 0:
Exception state (sv=0x3d058a00)
PC=0x0006E600; MSR=0x00009030; DAR=0x00000010; DSISR=0x40000000; LR=0x0006E584; R1=0x2F956780; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x0006E584 0x00089634 0x00367D00 0x00367D6C 0x007B62F8 0x007B71F8
0x007A9674 0x0037BCAC 0x007A5F64 0x00378F2C 0x00093FA4 0x0002C1C8 0x00024C94 0x000B2DCC
0x00000000
Kernel loadable modules in backtrace (with dependencies):
com.apple.ATIRadeon8500(5.2.8)@0x79d000->0x7e0fff
dependency: com.apple.iokit.IONDRVSupport(1.5.2)@0x785000
dependency: com.apple.iokit.IOPCIFamily(2.4.1)@0x5c4000
dependency: com.apple.iokit.IOGraphicsFamily(1.5.2)@0x75f000
Proceeding back via exception chain:
Exception state (sv=0x3d058a00)
previously dumped as "Latest" state. skipping...
Exception state (sv=0x3d06a280)
PC=0x9286C438; MSR=0x0200F030; DAR=0x23DBDA08; DSISR=0x0A000000; LR=0x92873360; R1=0xBFFEC540; XCP=0x00000030 (0xC00 - System call)

BSD process name corresponding to current thread: WindowServer

Mac OS version:
9D34

Kernel version:
Darwin Kernel Version 9.3.0: Fri May 23 00:51:20 PDT 2008; root:xnu-1228.5.18~1/RELEASE_PPC
System model name: PowerMac3,6
panic(cpu 0 caller 0xFFFF0003): 0x300 - Data access
Latest stack backtrace for cpu 0:
Backtrace:
0x0009B498 0x0009BE3C 0x00029DD8 0x000AF210 0x000B2A78
Proceeding back via exception chain:
Exception state (sv=0x3d058a00)
PC=0x0006E600; MSR=0x00009030; DAR=0x00000010; DSISR=0x40000000; LR=0x0006E584; R1=0x2F956780; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x0006E584 0x00089634 0x00367D00 0x00367D6C 0x007B62F8 0x007B71F8
0x007A9674 0x0037BCAC 0x007A5F64 0x00378F2C 0x00093FA4 0x0002C1C8 0x00024C94 0x000B2DCC
0x00000000
Kernel loadable modules in backtrace (with dependencies):
com.apple.ATIRadeon8500(5.2.8)@0x79d000->0x7e0fff
dependency: com.apple.iokit.IONDRVSupport(1.5.2)@0x785000
dependency: com.apple.iokit.IOPCIFamily(2.4.1)@0x5c4000
dependency: com.apple.iokit.IOGraphicsFamily(1.5.2)@0x75f000
Exception state (sv=0x3d06a280)
PC=0x9286C430

m1stake
Jun 27, 2008, 06:21 PM
Could just be going bad in general; mine is (no panics yet though).

Tracer
Jun 27, 2008, 06:24 PM
As opposed to Rember, try using the Diagnostics off of the Apple Hardware Test CD.

I had a similar problem with an MDD and it turned out that one stick failed about 1/3 of the time causing the kernel panics.

Tracer

Chromako
Jun 27, 2008, 07:41 PM
My dual 1.25 is really picky with memory sticks... sometimes it gets bored of them and decides one won't work any more, but Remember wouldn't find anything wrong. I heard somewhere that the MDD G4's were really picky with memory? Don't know if that helped...

shaneb
Jun 27, 2008, 10:17 PM
Another panic. And I already tried rember AND the AHT. That's how I found the stick I thought was bad.

I would have thought there would be someone here who could make sense of these reports... but maybe not.




Fri Jun 27 22:07:05 2008


Unresolved kernel trap(cpu 0): 0x300 - Data access DAR=0x0000000000000010 PC=0x000000000006E600
Latest crash info for cpu 0:
Exception state (sv=0x44eaca00)
PC=0x0006E600; MSR=0x00009030; DAR=0x00000010; DSISR=0x40000000; LR=0x0006E584; R1=0x2F926780; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x0006E584 0x00089634 0x00367D00 0x00367D6C 0x007B62F8 0x007B71F8
0x007A9674 0x0037BCAC 0x007A5F64 0x00378F2C 0x00093FA4 0x0002C1C8 0x00024C94 0x000B2DCC
0x00000000
Kernel loadable modules in backtrace (with dependencies):
com.apple.ATIRadeon8500(5.2.8)@0x79d000->0x7e0fff
dependency: com.apple.iokit.IONDRVSupport(1.5.2)@0x785000
dependency: com.apple.iokit.IOPCIFamily(2.4.1)@0x5c4000
dependency: com.apple.iokit.IOGraphicsFamily(1.5.2)@0x75f000
Proceeding back via exception chain:
Exception state (sv=0x44eaca00)
previously dumped as "Latest" state. skipping...
Exception state (sv=0x44e41c80)
PC=0x9286C438; MSR=0x0200F030; DAR=0xE00DD000; DSISR=0x42000000; LR=0x92873360; R1=0xBFFFCDF0; XCP=0x00000030 (0xC00 - System call)

BSD process name corresponding to current thread: WindowServer

Mac OS version:
9D34

Kernel version:
Darwin Kernel Version 9.3.0: Fri May 23 00:51:20 PDT 2008; root:xnu-1228.5.18~1/RELEASE_PPC
System model name: PowerMac3,6
panic(cpu 0 caller 0xFFFF0003): 0x300 - Data access
Latest stack backtrace for cpu 0:
Backtrace:
0x0009B498 0x0009BE3C 0x00029DD8 0x000AF210 0x000B2A78
Proceeding back via exception chain:
Exception state (sv=0x44eaca00)
PC=0x0006E600; MSR=0x00009030; DAR=0x00000010; DSISR=0x40000000; LR=0x0006E584; R1=0x2F926780; XCP=0x0000000C (0x300 - Data access)
Backtrace:
0x0006E584 0x00089634 0x00367D00 0x00367D6C 0x007B62F8 0x007B71F8
0x007A9674 0x0037BCAC 0x007A5F64 0x00378F2C 0x00093FA4 0x0002C1C8 0x00024C94 0x000B2DCC
0x00000000
Kernel loadable modules in backtrace (with dependencies):
com.apple.ATIRadeon8500(5.2.8)@0x79d000->0x7e0fff
dependency: com.apple.iokit.IONDRVSupport(1.5.2)@0x785000
dependency: com.apple.iokit.IOPCIFamily(2.4.1)@0x5c4000
dependency: com.apple.iokit.IOGraphicsFamily(1.5.2)@0x75f000
Exception state (sv=0x44e41c80)
PC=0x9286C430

shaneb
Jun 27, 2008, 10:21 PM
My dual 1.25 is really picky with memory sticks... sometimes it gets bored of them and decides one won't work any more, but Remember wouldn't find anything wrong. I heard somewhere that the MDD G4's were really picky with memory? Don't know if that helped...
Oh, and as a side note... the one it decided it didnt like was a factory 512 chip.

Chromako
Jun 27, 2008, 10:36 PM
One of the rejected sticks I had was a factory one, too... alas, out of warranty.

shaneb
Jun 27, 2008, 11:58 PM
One of the rejected sticks I had was a factory one, too... alas, out of warranty.

Ha, yeah, way out of warranty. I just ran AHT again, no problems. Then ran DiskWarrior again, some minor problems, but nothing major. If I continue to have problems I guess I will transfer all my data and create a new admin account and try that. It worked for me once before when I had constant KPs on a machine at work.

cal6n
Jun 28, 2008, 05:40 AM
My old dual 1.25 used to kernel panic like crazy. I thought it was the RAM but it turned out to be the RAM slots themselves.

After I bought a can of servisol aeroklene 50 (a specialist electronics cleaning solvent; DO NOT use an ordinary switch cleaner!) and cleaned each slot, re-seating each RAM stick repeatedly with some enthusiasm, my problem was entirely cured. That was in 2003 and it still works well as my mate's main music production machine to this day!

shaneb
Jun 28, 2008, 11:47 AM
My old dual 1.25 used to kernel panic like crazy. I thought it was the RAM but it turned out to be the RAM slots themselves.

After I bought a can of servisol aeroklene 50 (a specialist electronics cleaning solvent; DO NOT use an ordinary switch cleaner!) and cleaned each slot, re-seating each RAM stick repeatedly with some enthusiasm, my problem was entirely cured. That was in 2003 and it still works well as my mate's main music production machine to this day!

Hey thanks! I'll add that to my list of things to try.

willrunz
Jul 1, 2008, 12:11 AM
I have a G4 FW800 1.42dp that has had problems with endemic kernel panics for months now. Trouble shooting has been very difficult as it has not once written a panic.log file. I have run every software test known to mac and it always comes up clean. My confidence was already wary as most of what I have read has informed me that this is insufficient. However, I think I may have figured it out today and wanted to share my hypothesis.

My rig was booted up around 10:30am. I began working and was pleasantly surprised that I had no issues through most of the morning. However, by noon, I could not work for longer than 20 minutes with out a kernel panic. Embarassed, as I had a client over, I felt the need to do something even if it was just for show. I opened my mac and pushed on all the pci cards and ram dimms to make sure they were seated. I closed the mac and started working again.

True to form, it crashed as soon as I got Protools open. My client and I had a little talk about what I had done previously to trouble shoot the problem. It was then that I thought about running the mac while it was open. I opened it up, rebooted and started up protools. 30 minutes went by with out a problem. next time I looked, 2 hours had passed with out a crash. I went the rest of the day with out any problems.

It was then that I noticed something interesting. I have four hard drives installed. The heat sink for the processors, when the mac is closed, basically sits on top of one of my internal work drives. Could my kernel panics be a circuit thermal shutdown? with the lid open the heat sink is exposed to room temperature air. while closed it is having to dissipate both the heat from the chips and the heat from the drive.

I once witnessed an external fw drive go into a similar state. shut it down and let it cool of and its fine. This is what I suspect is wrong with my machine. Its a design flaw that only manifests in non-ideal situations. if I had ac or maybe one less hard drive, I may never have seen this problem manifest.

So before you send it in for the motherboard/ram/HD replacement and before you spend hours running dumb tests that don't tell you squat, try running the computer with and extra fan or while open. Can't hurt right?

shaneb
Jul 14, 2008, 09:43 AM
An update to 10.5.4 from 10.5.3 seems to have fixed the problem. 7.5 days uptime with no problems at all.

http://discussions.apple.com/thread.jspa?threadID=1587691&tstart=0