Apple's help forums don't seem to have anything useful for me, so here we go. Somebody here must know something useful.

Background: my home computer is a PowerMac G4, dual 1.0GHz processors, with 1GB of RAM. This is one of the models with dual CD slots, rather than the newer "mirror" ones that, among other things, have a better memory architecture.

Anyway, I like the idea that my computer can go to sleep by itself so I have the feature enabled. It saves power, and it also makes a whole lot less noise when it's asleep. Recently (starting around OS X, v10.2.4 or so), it's been having trouble waking up. I'll see my old screen, which will promptly go grey and give me a dialog box telling me I have to reboot my computer by holding down the power button. Yummy.

There's nothing particularly revealing in the system log:

Jul 4 16:10:01 DapperMac CRON[805]: (root) CMD ( /sw/sbin/anacron -s)
Jul 4 16:10:01 DapperMac anacron[806]: Anacron 2.3 started on 2003-07-04
Jul 4 16:10:01 DapperMac anacron[806]: Normal exit (0 jobs run)
Jul 4 16:43:29 DapperMac syslogd: restart
Jul 4 16:43:29 DapperMac mach_kernel: standard timeslicing quantum is 10000 us
Jul 4 16:43:29 DapperMac mach_kernel: vm_page_bootstrap: 124948 free pages
...
Jul 4 16:43:30 DapperMac /usr/libexec/panicdump: Error (-1) setting variable - 'aapl,panic-info'
Jul 4 16:43:30 DapperMac SystemStarter: crash reporter (244) did not complete successfully.
Jul 4 16:43:30 DapperMac root: setting hostname to DapperMac.local.

Presumably, the cron job is what put the computer to sleep. Then, the next log entry is from the machine rebooting, and there's apparently no crash dump of any kind to indicate what went wrong.

This behavior happens maybe once per week. It's often enough that's it's annoying, but it's not 100% repeatable. The computer is basically stock. I've added a second hard drive, but that's it. Of course, there's a ton of stuff hanging off the USB ports. Should I try unplugging all that stuff? The only recent addition that might possibly be related is my Onkyo SEU-55 USB sound card. I've noticed that playback through the external sound card has pops and clicks, while the internal sound output is clean. I have no special drivers installed; it's just using Apple's built-in USB audio drivers.


Then, once we solve this problem, maybe we can figure out why wake-on-LAN isn't working. But, one thing at a time...

It wouldn't hurt to unplug all the USB goodies and see if that helps. I have had similar problems with my (much older) mac for a couple of reasons:

1. Old PCI cards that aren't compliant with the power managment. Sounds like this probably isn't your issue.
2. Overheating. My beige G3 will overheat and crash while asleep if the ambient temp is above 85-90 degrees. My box is very loaded (3 PCI cards, 3 hard drives, etc) so this isn't that surprising. You might want to try blowing out the inside of your case and power supply with some canned air and see if that makes any difference.

Just a couple guesses.
-Mike

The screen you are seeing is the pretty kernel panic screen, so there is a good chance a kernel panic log has been created. Open Apple System Profiler, and check the Logs tab for a Kernel Crash entry. Posting the details here might help. Also, post the details at macnn.com, as some people there tend to run the kernel dumps through an analyzer and can give you a decent explaination.

Okay, here's what the crash log says:

Fri Jul 4 16:43:30 2003




Unresolved kernel trap(cpu 1): 0x700 - Program DAR=0xe14ab000 PC=0x02a1ab10
Latest crash info for cpu 1:
Exception state (sv=0x1512EC80)
PC=0x02A1AB10; MSR=0x00089030; DAR=0xE14AB000; DSISR=0x42000000; LR=0x02A1AB10; R1=0x0D91BBF0; XCP=0x0000001C (0x700 - Program)
Backtrace:
0x02A1AB10 0x02090800
backtrace terminated - unaligned frame address: 0x0000FFFF

Proceeding back via exception chain:
Exception state (sv=0x1512EC80)
previously dumped as "Latest" state. skipping...
Exception state (sv=0x14E62280)
PC=0x00000000; MSR=0x0000D030; DAR=0x00000000; DSISR=0x00000000; LR=0x00000000; R1=0x00000000; XCP=0x00000000 (Unknown)

Kernel version:
Darwin Kernel Version 6.6:
Thu May 1 21:48:54 PDT 2003; root:xnu/xnu-344.34.obj~1/RELEASE_PPC


panic(cpu 1): 0x700 - Program
Latest stack backtrace for cpu 1:
Backtrace:
0x0008581C 0x00085C4C 0x000287B4 0x0008F6C8 0x000927D8
Proceeding back via exception chain:
Exception state (sv=0x1512EC80)
PC=0x02A1AB10; MSR=0x00089030; DAR=0xE14AB000; DSISR=0x42000000; LR=0x02A1AB10; R1=0x0D91BBF0; XCP=0x0000001C (0x700 - Program)
Backtrace:
0x02A1AB10 0x02090800
backtrace terminated - unaligned frame address: 0x0000FFFF

Exception state (sv=0x14E62280)
PC=0x00000000; MSR=0x0000D030; DAR=0x00000000; DSISR=0x00000000; LR=0x00000000; R1=0x00000000; XCP=0x00000000 (Unknown)

Kernel version:
Darwin Kernel Version 6.6:
Thu May 1 21:48:54 PDT 2003; root:xnu/xnu-344.34.obj~1/RELEASE_PPC




And, there's a whole slew of other ones just like it. This doesn't say much of anything useful to me. Can any of you make sense out of it, or should I go post on MacNN?

I'd try the suggestions in this thread. Specificially "Try deleting this file:
/Library/Preferences/com.apple.PowerManagement.plist".

If not, go post a detailed thread there and see if someone can figure out exactly what it is. All the kernel panics that I have had have been easy to debug, due to a single program causing them. Yours looks like it may be a true panic starting in the kernel.

Okay, I deleted that file. Interestingly, it was dated June 11, 2002. When I tweaked my settings in the System Preferences, that file didn't change, but /var/db/SystemConfiguration/com.apple.PowerManagement.xml reflects the current values.

Fun, fun. We'll see if the computer crashes again.

This keeps getting weirder. My previous crash, that led me to post here was on July 4. Now, one week later, it crashed again with exactly the same symptoms. This time, however, there was no kernel panic log entry at all. Nothing.

More disturbingly, check out this from the system logs. Here's what it's supposed to look like when you put the machine to sleep:

Jul 9 22:32:11 DapperMac mach_kernel: AppleNMI mask NMI
Jul 9 22:32:11 DapperMac configd[112]: executing /System/Library/SystemConfigur
ation/Kicker.bundle/Resources/restart-AppleTalk
Jul 9 22:32:13 DapperMac mach_kernel: System Sleep
Jul 10 07:42:49 DapperMac mach_kernel: System Wake
Jul 10 07:42:49 DapperMac mach_kernel: Wake event 0020
Jul 10 07:42:49 DapperMac mach_kernel: enableClockSpreading returned with 0
Jul 10 07:42:49 DapperMac mach_kernel: AppleNMI unmask NMI
Jul 10 07:42:49 DapperMac mach_kernel: UniNEnet::monitorLinkStatus - Link is up
at 100 Mbps - Full Duplex

Now, here's what I see before the crash:

Jul 10 07:49:06 DapperMac mach_kernel: AppleNMI mask NMI
Jul 11 10:21:58 DapperMac syslogd: restart
Jul 11 10:21:58 DapperMac mach_kernel: standard timeslicing quantum is 10000 us
Jul 11 10:21:58 DapperMac mach_kernel: vm_page_bootstrap: 124948 free pages
...


So, it looks like the machine never went to sleep "properly", causing it to freak out when it woke back up again. Further disturbing is this:

Jul 11 10:22:15 DapperMac SystemStarter: Waiting for local disks
Jul 11 10:22:48 DapperMac last message repeated 11 times
Jul 11 10:24:51 DapperMac last message repeated 41 times
Jul 11 10:28:21 DapperMac last message repeated 70 times
Jul 11 10:28:22 DapperMac /sbin/autodiskmount[186]: Mounted ufs /dev/disk0s2 ...

What's going on when it's waiting for the disks? Was it fsck'ing? I grepped through all my prior system logs, going back to the beginning, and I never saw this message before. Very disturbing.

At a wild guess, I'd say that the disks got powered down but the OS/computer never properly did.

Do you use AppleTalk? If not, and it's on, try turning it off. (I haven't used AppleTalk since well before OSX, so I'm not really sure that even makes sense, but try it anyway.)

Note that I'm talking wild guesses. I could be completely off base.

All my illusions are shattered.
I thought Macs NEVER crashed. Let alone gave messages as readable as this one!

Does that mean the the OS X Macs are just as likely to do the single mouse version of the BSOD?

At a wild guess, I'd say that the disks got powered down but the OS/computer never properly did.

Take a look again, if I am reading that right it took almost 6 minutes to spin up the disk. Sounds like bad hardware to me. Anyway, if the drive routinely takes that long to spin up maybe that is what is causing the crash when it wakes up. Have you tried doing a drive check by hand? Boot single user by holding down 's' when the computer starts up and when you get a prompt do '/sbin/fsck -f /'.

-Mike

My reference point was that the failed one never reached mach_kernel: System Sleep.

But your point is wildly valid, too. Maybe they're interrelated.