Q: Dual Quad-core Mac Pro reboots from sleep

That is exactly why I refuse to take my MP to an Apple Store for service. It's very likely that when they plug it in and test it, it will work (because of the SMC reset as you mentioned). They'll then say "it's fine" and tell me it must be a "software issue" and to do another re-install of Leopard bla bla bla. A complete waste of time.

The new MPs do in fact have an SMC reset switch, located just right of center behind the top RAM riser.

I was "startled" when I read there was none. Glad to see it is there. Now to get someone to do a nice photo of the little devil!

Prior system had to remove bay #4 to access the switch.

I think you could say every system has to have PMU/SMU or SMC of some sort.

http://www.macitynet.it/immagini/visitaconnoi/wwdcsf06/macpro/source/8.shtml

So what did I do? I turned off the MP, let it cool down for ten minutes or so, then unplugged everything starting with the power cord, then DVI and then USB and Firewire cables. After opening up the case, I then removed the top RAM riser and used a wooden dowel (non-conductive) to press the SMC_RST briefly just until the LEDs to the right of it lit up, (a second or so). Put everything back together and plugged all the cables in leaving the power cord for last. I then hit the power button and did a PRAM reset, letting the unit chime 4 times - read somewhere despite the recommended 2 that you should always let it go for 4 times. I figured what the heck. So far so good as I've said before. Since that time I have upgraded to the 8800gt and still haven't had the issue return.

This is very interesting. I'll try it and report back - nothing to lose, after all.

Try to find that little button. I know on mine that there IS a major difference between PRAM and unplugging the system, and actually using the SMC reset. I had to use the button when my Pioneer SuperDrive stopped being recognized once. Could have been after an SMC firmware update.

Good foresight. That's exactly what happened to me. I took my machine in and it worked flawlessly in the shop and for about a week after I returned. Apple needs to acknowledge this issue and work toward a solution. I hope everyone has reported this issue on the feedback page. As others have pointed out, Apple does not pay attention to or even cares about what is said in this discussion group.

http://www.apple.com/feedback/macpro.html

I have posted a picture of the SMC_RST that I just took of my MP 2008 3.0. As I said before it's just behind the top RAM riser on the right.

http://www.flickr.com/photos/24568684@N07/2326213307/

That is easier to distinguish and very clear job! I'd hope this could be helpful to others having trouble. Thanks!

Your post is interesting. I have had both the reboot after sleep and the freeze where you can only move the mouse. I had fixed the reboot problem by doing the PRAM and SMC reset. I kept having the freeze problem and took my computer to the Apple store. They had it for a week and did not get it to act up, but did an archive and install just for grins. When I got it back, It had the reboot after sleep problem again. I did the SMC and PRAM reset, but only waited for two chimes because that was what I had read somewhere. Before I took the computer in, I always waited for three for some reason and the problem was fixed. I noticed that the reboot problem was not fixed after doing it for just two chimes. I reset the SMC and PRAM again, this time waiting for three chimes and it has seemed ok for a few days. For whatever reason, I have not had a freeze since getting the computer back, although it is still not long enough to really tell. I have not been running parallels or using my USB hub for the last few days either, but I don't know that that is related. I did the firmware update and have not had any problems from it. At this point I would definitely recommend waiting at least three chimes when doing the PRAM reset before you decide that the SMC/PRAM reset does not work.

The number of times to hold keys down to reset PRAM/NVRAM (which must be done from a cold boot, btw) has had some urban legend going back years to the old pre-G3s and OldWorld Open Firmware and longer.

Resetting NVRAM from Open Firmware was always preferable, but no longer an option now with EFI.

In the past, not all bytes were cleared or NVRAM device tree rebuilt by PRAM reset. Two should be already "one extra" - if it works. A 'warm' restart PRAM reset doesn't touch NVRAM.

SMC_RST button is ideal.

Over 450 posts in this thread, and no official word from Apple yet?

Somebody at Apple has to come out with a statement. This is getting silly.

All I can say is there are going to be a ton of 8 core refurbs available soon.

What's the word Apple? Wait for a patch/update or return the hardware?


JT

Why? planning to replace MacBook Pro? Firmware updates like the ATI yesterday tend to take time.

Apple never posts more than "to increase system stability" or compatibility, as they did with some iWork updates for Leopard, nothing specific so even though there were problems, it was "try this" and see if it help. So I'd guess this falls under something similar if at all.

Right now, resetting the SMC is about the only thing.

I assume that there may be different motherboards being shipped, too.

I don't mind the time it takes for a Firmware update to come out. I get that part.
The problem for me is the lack of acknowledgement that there is an issue. If Apple could just say, hey, we know about this and we are working on a fix/patch/update it would alleviate allot of stress and worry that myself and my clients have wondering if it is something about the individual build. Same thing for the ATI-2600 update, until the time the firmware update came out, the standard answer from Apple tech support was to format and reinstall the OS. That gets a little difficult when you spend a day building one of these boxes as a creative production machine. Apps/Fonts/Updates plus configuration.
Blowing this all away when it's a hardware or driver issue is really a painful process.





JT

I'm with you, JT. If it does turn out to be a hardware problem, I don't want to be stuck with one of those boxes. I know there is technically a one-year warranty on hardware stuff like this, but my experience with this kind of thing w/Apple in the past hasn't been positive. I'm nervous because they still haven't acknowledged it, which basically means it's OUR problem and not THEIR problem. An acknowledgment shifts the burden of responsibility from us to them. That's why it's important.

Now that's interesting. I have that button, but it's labeled "DIAG" (just above the button), not SMC_RST.
To the right of the button is a row of SMD LEDs labeled (L-R): SLP, STBY, ERRB, ERRA, OT B, OT A, GPU, PWRG, and EFI.

The Apple part number that also appears in the photo...820-2128-A...is the same on my board.

I also have an unlabeled pushbutton (same type) just below drive bay 4. Just to the right of that switch, and labeled "RST" is a pair of what would appear to be test points, about 0.1" apart vertically. A trace apparently leads from the top one to the pushbutton.

Hmmm. Are you having the sleep issue as well? (I imagine you are, since you're following this thread).

To the original poster: do you have all of the same buttons that BillC37 has, with the same labels? Does anyone know what the "DIAG" label means?

This is confusing!