Bear with me here for a second...

I have an early 2008 MacPro3,1 that has been my work horse for many years and for the last several years has been running continuously as the home server. I've noticed in the last week that if I bang the desk that it is on, the machine will cold reboot immediately. (how did I discover it? hitting a stapler for big stack of papers)

Machine hasn't moved in quite some time and wasn't sensitive like this until perhaps the last week. Everything should be buttoned down well in the machine except perhaps the Bluetooth adapter on the internal mini PCI port, but the WiFi/Bluetooth is working just fine. Also, power cord is snug.

So I suppose the question is whether or not there is something in the PSU that might be sensitive to movement or perhaps something else in the machine? Other than this, everything else seems fine with the machine.

UPDATE 2017-03-14: One other thing to note is that when it reboots, I get the startup chime followed by ~10 seconds of no activity and then another startup chime and then it goes into the full boot routine. Almost feels like a CMOS reset or something like that, i.e. is there a reset button on the motherboard that perhaps is a bit twitchy? (no, I haven't re-seated things yet since I haven't had the time, but I did open the case to check that the two cards, the WiFi/BT module and memory risers are snug).

  • 1
    Just to clarify, since discovering this, have you opened the Mac Pro and taken a look inside to check for anything that may be amiss? – Monomeeth Mar 13 '17 at 1:19
  • Can't post an answer as I have nothing but anecdote - I once knocked into my 3,1 whilst hoovering [quite gently, I thought at the time] & the same happened. That was maybe 3 years ago & I've just made sure to never do it again. The machine has had a full strip & clean since then so maybe I've fixed whatever it was, as it still gets hoovered round [gently] every week. – Tetsujin Mar 13 '17 at 7:50

This something is quite possibly a loose PCI card, maybe your video card, it can have some stress on it from the weight and pull of the VGA or DVI cable. Could be a usb daughter card as well. Unplug the machine from the power mains. Reseat the cards, blow the dust off them. Maybe it won't reboot on bump any more.

tl;dr: A loose PCI card can cause a reboot, they aren't hot-swappable.
  • A spinning HD from 2008 may work just fine in a stationary machine, but the bearings can lose tolerance for extreme motions of the spindle exacerbated by sharp movements or shock. In addition to reseating the PCI board(s), check to make sure RAM is firmly seated in its carriers, and that excess dust has been removed from the inside. I've worked on many MPs of this vintage, and as solid as they are, they need a little TLC and updating. – IconDaemon Mar 13 '17 at 1:50
  • I'll accept this one as the answer since I seem to have solved the problem by re-seating the memory risers (and blowing some dust out of them, but I'm sure that has nothing to do with it) and I did wiggle around the video card a bit (but didn't re-seat). Seems to have cleared up. – bjb Mar 17 '17 at 11:49

Possibly faulty connectors in the back. Connected devices that are faulty come to mind too. I don’t want to say it’s the board, but in my experience, it could be if it’s nothing else.


Greetings and thanks for this thread. Apologies for jumping in but I have also had this problem recently with my MP 3,1. Just small knocks on the desk cause a reboot after which it fails repeatedly to start and i have to force it off with the power button. Is there something loose inside? Power supply? I can't find out.

Probably as a result of these episodes, I've lost all my RAM on one riser (but the other riser's RAM is left untouched). This has happened twice now and the first time it corrupted a HD as well. The second time (today) I left it for a few hours and managed to get it to start up ok, when I checked and found I'd lost 4x2GB of RAM. But I'm not so sure the problem is related to heat because it occurred this morning from a cold boot.

Naturally I am worried to get replacements in case it happens again. It seems the reboots and the RAM failures are connected - but how?

I have explored for answers, including replacing the power supply unit and replacing the GPU (Stock ATI Radeon HD2600) and I've speeded up the fans. If anyone else could give a few pointers at least as to how to check things like the GPU, PSU, etc, and what might be causing the RAM to fail -I'd be eternally grateful.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .