Dual Quad-core Mac Pro reboots from sleep

Tom -- DeepSleep seems (for now) a reasonable solution to the rebooting problem on my 2008 MacPro. DeepSleep seems to function as it should, these models. Sure seems to be deep-sleeping.

Hopefully the situation is being redressed (the tech support people say this is so), then we can go back to using our systems in normal fashion.

In the Energy Saver control panel under "Options", is "Restart automatically after a power failure" checked or unchecked for those having problems?

Mine has always been unchecked and I have "The Problem" !!

According to reports on AccelerateYourMac and here, it doesn't matter what setting is used for restart; whether you have 1500VA or none. Only resetting PRAM/NVRAM from cold starts and SMC reset which can be temporary.

Has anybody gotten anything definitive from Apple about what they think the problem is? i.e. software, firmware? Do they say the problem has definitely been noted and a fix is in the works? etc. Something like that.

If it was a software problem, why wouldn't everyone be experiencing it? According to a poll on MacRumors, only 1 out of 5 are having this issue (any issue, actually) with their Mac Pro '08s.

If it's a hardware problem that affects only some systems, that could explain why it's not widespread. If this is true, then it's also true that Apple hasn't fixed it yet or isn't even aware of what's causing it, because people who get new machines have the same problem.

Of course I could be missing something here. Is there some way of explaining how a core system software problem would only affect some machines, and not others?

Has anyone actually received a working replacement Mac Pro?

It would be helpful to summarize everything we know so far.

Maybe a lot of owner's have the problem and are not reporting it. Or that many are not using the sleep mode. Or when the problem occurs, they just figured it was something they did wrong, etc. Or possibly the one's without the problem stumbled upon certain settings that prevent the issue from occurring.

It certainly will be interesting to see what the final diagnosis is.

Over at MacFixIt there are many reader reports about the MP '08. Plenty of people clearly state that they don't have the "reboot from sleep" issue. So it's evident that not everyone is experiencing this. If they were, there would be many more threads here and this one would be even longer.

I've read that people running Windows (native) don't have the problem. If it's true that there isn't a single report of this, then that would strongly indicate a software/firmware problem. Does anyone have more information on this?

My mac have the sleep problem, i zapped the pram and now after 2 long sleeps it did not reboot, i'll try more theese days.

I have an ups and i turn it off every night, so the mac remains without power every night, probably this helps.

Mine has only rebooted from sleep three times, and I'm sleeping it quite a bit while experimenting. I answered the poll on Mac Forums before I realized that I also had the problem, so others may have done the same thing, and then you can't change it. If some owners don't sleep their computer much, they might not realize that there even is a problem, because otherwise, the computer works great. A couple also reported that their replacements were doing the same thing.

I have the same problem, i try to reset SMC and PRAM/NVRAM.

i see that restore from sleep mode is ok only when parallels desktop not running or PD is not running before.
if i run PD and close i see the same problem, return from sleep restart the machine.

i must see if is a problem of PD or problem of ram occupated when i stop my mac, because i use PD with 1 GB of ram for the virtual machine.

next :

i open 20 application in my mac i see about 2 gb of ram active, i try to stop and resume and OPS !!
....RESTART !!!

IS A RAM PROBLEM ????

JimRobertson wrote:
I don't have a UPS, and I'd be grateful if you could tell us how much power your machine draws under several situations; e.g.,

1. "awake," monitor(s) on, just sitting there waiting for the user to do something
2. monitor(s) black, hard drive(s) spun down, but entire system NOT set to sleep.
3."real" sleep (you've already stated that datapoint - 36 watts - but I list that just for comparison.
4. During a workout (something that "pins the meter" on a few of the processor cores).

Won't solve any of our problems, but it will give us all some food for thought about what to do for now.

Thanks so much,

Sure...this is of course off the topic of this thread, but here's what I'm seeing:

Sleep: 35 Watts
Awake, but monitor sleeping and disks spun down: 108-110 W
Idle with 2 x 2 GB RAM installed: 195-199 W
TM backup as only activity: 204 W
Rendering with all 8 cores pegged at 100%: 240-272 W

Idle with Apple 2 x 1 GB RAM only installed: 175-180 W

Back to the main topic of the thread:

Based on this macfixit article...
http://www.macfixit.com/article.php?story=20070430081644196

I'm slowly trying pairs of memory cards to try to find the problem with this reboot from sleep and other problems. With 4 x 2 GB OWC and the 1 x 1 GB Apple (10 GB total), I had the problem. With just the Apple RAM I could not reproduce the issue. Called OWC and they confirmed that it is possible for RAM to pass all tests and still cause strange issues. So, now trying with each pair of OWC RAM, then will try with all four OWC, etc. until I can get the problem to appear again. I still have a hunch that it may not be RAM, but rather the Mac Pro's power supply not providing consistent power to all of the RAM during sleep, but I have no basis for that.

Frankly, if RAM passes all tests and yet can still cause problems, to me that means that Apple's hardware/software is not robust as it could be. Component failure should be dealt with gracefully - and if a component doesn't fail any tests, it is hard to call it a failure - but a mismanagement of some kind of tolerance, signal, voltage or whatever by Apple.

Testing continues...

Karl

Please keep us posted on your RAM experiment. When I removed all of my 3rd party (TransIntl) RAM, I couldn't reproduce the "reboot from sleep" problem either.

I put all of my RAM back in, and began to have the problem again. The reason I haven't tried testing yet is that 1) many people have the problem even with stock Apple RAM, and 2) I can't get by with 2 GB of RAM due to intensive graphics/photography work I do.

If your RAM experiment works, though, I'll have to bite the bullet and try it.

Back off topic: This power consumption thing is interesting, so glad you asked.

I use the 'Beach' screensaver (Ken Burns mode)...and just noticed that the computer uses 20 watts more power with this running than with just the desktop displayed.

Kind of brings a new concept to the cost of running an app, too. Run Grapher and from Examples select Conchoid. No increase in power for me over idle machine. Select Torus-Knot from Examples - and it takes about 27 watts extra.

Karl