Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Shadowkahn

macrumors newbie
Original poster
Sep 1, 2009
19
0
Recently came back from an extended trip in Michigan to my home in Florida, only to discover that my Mac Pro is randomly restarting both with my Windows Partition and Mac Hard Drive. Restarts seem to be occur quicker when playing a Windows Games, or doing some heavy loads while on the Mac side - though it can and will happen if I play some World of Warcraft on the mac. The computer restarts, no kernel panic, just simply restarts. On the Mac side of things it will relaunch the application I was running (even though I have setup Lion to not do this).

System has a brand new AMD HD5770 installed, I have cleared out the dust in the system, though not going so far as changing the thermal paste on the CPUs/Heatsink - overall the system should be fine. I suspect that the issue is the PSU, but it being the Holidays I'm trying to save money instead of wasting it to take this system to the local third party Apple approved repair shop. The question I have is this - am I correct to think this is the PSU needing to be replaced? Is sudden restarts a symptom? Or is this an overheating issue? I have run Apple Diagnostic Disk multiple times, not experiencing any of the random restarts that occur when actually using the system.
 
Do the programs close and the OS shutdown like a restart? Or does it just power off then instantly turn back on? If it's the latter, then I agree with the previous poster about it being the power supply.
 
That may very likely be a power supply - in this case it's bad...

But, it also can be RAM issue. In my case (Mac Pro 3.1) random restarts disappear after reinstalling all RAM. Actually that was sort of cleaning RAM contacts.
 
Sorry about taking so long to respond to this thread

Was out of town (once again) for an extended amount of time. While away I received a gift of 12gb of RAM from OWC. For last few days I have been running the computer solidly with just one riser with 4gb of ram (2x2gb ram) and the other riser empty. No restarts. The issue that I was having with restarts is exactly like a sudden shut down and then turning itself back on (power off and power on). But only having one riser in seemed to resolve that issue. So when I got my gift today from my friend who bought 12gb of OWC I thought I would test to see if I was going to have an issue putting ram into the computer.

Once filling both risers (since I do have 4gb of Ram already) and slapping both risers into the Mac, I turn the computer on. And a minute does not go by before the computer shuts down and starts up again.

Going against the Apple PDF of installing Ram in the computer, I empted one riser of ram and left 8gb on the other (2x2gb and 2x2gb), and the computer runs solidly. But if I even put just 12gbs of Ram (as in filling one riser with 8gbs of ram and the other with just 2gb) the symptom of power off and power on returns.

Currently I am typing on the machine in question and with just 8gbs on one riser seems to be running fine with no sudden cutoff of power and restarts both from within the Mac OSX and Windows 8.1 operating systems.

As a test I installed the previous graphic card and the restarts still occur when both risers have RAM occupying all memory slots. If one is left empty (and I am using different riser slots and riser cards while testing this) and the other is full - the computer works fine.
 
Was out of town (once again) for an extended amount of time. While away I received a gift of 12gb of RAM from OWC. For last few days I have been running the computer solidly with just one riser with 4gb of ram (2x2gb ram) and the other riser empty. No restarts. The issue that I was having with restarts is exactly like a sudden shut down and then turning itself back on (power off and power on). But only having one riser in seemed to resolve that issue. So when I got my gift today from my friend who bought 12gb of OWC I thought I would test to see if I was going to have an issue putting ram into the computer.

Once filling both risers (since I do have 4gb of Ram already) and slapping both risers into the Mac, I turn the computer on. And a minute does not go by before the computer shuts down and starts up again.

Going against the Apple PDF of installing Ram in the computer, I empted one riser of ram and left 8gb on the other (2x2gb and 2x2gb), and the computer runs solidly. But if I even put just 12gbs of Ram (as in filling one riser with 8gbs of ram and the other with just 2gb) the symptom of power off and power on returns.

Currently I am typing on the machine in question and with just 8gbs on one riser seems to be running fine with no sudden cutoff of power and restarts both from within the Mac OSX and Windows 8.1 operating systems.

As a test I installed the previous graphic card and the restarts still occur when both risers have RAM occupying all memory slots. If one is left empty (and I am using different riser slots and riser cards while testing this) and the other is full - the computer works fine.

This is exactly what happen to my 2.1(also with a 5770). Random restart and only recognize part of the memory(it shows only Hynix but no OWC DIMMs). The led on the riser card flash when I install OWC memory(bed RAM signal). Don't know if running OWC modules ruined the lower raiser cards, on the upper card I've different brand and everything it's fine. Anyway I filled up the upper riser with all the RAM that I can leaving the lower riser empty and the problem was gone for about 8 months.
Now it's starting to restart again, but not randomly.. This time was a different problem, it happen constantly only when I'm running all 8 core at full load. Temperature rise around 100C in about 10/20 minutes, then the mac restart. If I render with only one or two core obviously the temperatures are lower and the computer run just fine. My guess is bed thermal paste(it was somehow confirmed 8 months ago by Apple assistance diagnose tools reporting CPU proximity problem). I'm not going to replace thermal paste because I'll have my new computer in less than a month. In the mean while for a "temporary" solution I've installed smc fan control utility, increased CPU fan speed to 1500rpm(before it was spinning only at no more than 500 according to hardware monitor, strange..). By increasing fan speed CPUs temperatures are now at 50/60C and that solved 100% the problem.
 
Last edited:
Well the problem is not ram not being recognized, ram is seen initially before it begins the shutdown and turn back on. And as stated before the risers (either one) can be filled to capacity as long as one is left empty the computer will boot and work as normal. Now if I do fill both risers the issue arises.

If it's a simple matter of cleaning off the CPU thermal paste and reapplying new paste I will do that next week and report my findings. In the process of getting dust out of the system over the years I have been able to remove the riser cage and the front fan assembly with ease.
 
Well changing out the thermal paste didn't change anything. I feel better about my Xeon processors cooling but the same sudden power off and restart occurs when using both risers fully packed with ram, or slowly filling them with two dimms at a time. If the second riser has any ram installed (and interchanging risers does not change this issue) the computer will shutdown and restart. SO my best guess is the Mac Pro PSU. I'll probably take it in to a local shop to make sure before I drop 400 or more for a PSU. I'll try to keep everyone informed.
 
Sadly haven't had a chance to fully investigate - had to recently move. Have been told before I left that it could be the actual memory over heating but that is a stretch. So once I am settled here I will be replacing the PSU when I get a chance. Seems to be the only option other than replacing the logic board - which really if it comes to that point I will just get a new computer.
 
Without doubt it would make more sense to replace rather than repair if it involved the board, those boards still fetch crazy money, my model being the 3ghz 8 core, I still find it hard to believe the asking price is around the same as replacing the entire machine.

Just so you know, my problem is identical, all the ram works fine, either fully
Populating riser A or B in either the upper or lower ports, the problem arises when both riser boards are installed, it either fails to start, or on the rare chance it does fire up, shortly after it begins a restarting process and never actually restarts, it just gets stuck in a boot loop over and over.
The machine is immaculate inside, the entire guts of the machine have been taken out, carefully cleaned with air, and reseated. The SMC / PRAM have been reset countless times, everything I could suggest cleaning I've already done.
Replacing the PSU i have considered, however that's a huge gamble in something that may or may not solve the issue. Bare in mind, my machine usually powers 4 sata drives, one SSD drive, a 5770 card, 16gb of ram, and a SuperDrive, leaving me thinking maybe I was overloading it a little, however removing all the drives, the SuperDrive, and installing my old ati 1900 card back in there, surely I would have lightened the load... And yet the problem
Is still happening, doesn't point me in the direction of a PSU.

As things are right now, it's running stable with one riser, just a shame to sacrifice the RAM on the other card. Let's hope someone browsing these forums finds a solution why this is happening.
 
Without doubt it would make more sense to replace rather than repair if it involved the board, those boards still fetch crazy money, my model being the 3ghz 8 core, I still find it hard to believe the asking price is around the same as replacing the entire machine.

Just so you know, my problem is identical, all the ram works fine, either fully
Populating riser A or B in either the upper or lower ports, the problem arises when both riser boards are installed, it either fails to start, or on the rare chance it does fire up, shortly after it begins a restarting process and never actually restarts, it just gets stuck in a boot loop over and over.
The machine is immaculate inside, the entire guts of the machine have been taken out, carefully cleaned with air, and reseated. The SMC / PRAM have been reset countless times, everything I could suggest cleaning I've already done.
Replacing the PSU i have considered, however that's a huge gamble in something that may or may not solve the issue. Bare in mind, my machine usually powers 4 sata drives, one SSD drive, a 5770 card, 16gb of ram, and a SuperDrive, leaving me thinking maybe I was overloading it a little, however removing all the drives, the SuperDrive, and installing my old ati 1900 card back in there, surely I would have lightened the load... And yet the problem
Is still happening, doesn't point me in the direction of a PSU.

As things are right now, it's running stable with one riser, just a shame to sacrifice the RAM on the other card. Let's hope someone browsing these forums finds a solution why this is happening.

Hi Shibuyablue,

So far I have counted more than 700 Mac Pro users, who have made some form of post/reply across various mac forums and Apple Support Community who have experienced random restarts on Mac Pro's 2007-2012 models. This is appalling and a thorough investigation by apple should of been made resulting in a recall to access/repair/replace the faulty components(s). Why there is some who have not experienced the random restarts, which could be numerous and varying reasons, have saved from the annoying and time wasting efforts by the end user, and in many cases apple genius bar staff, trying to find the actual fault.

While iStat menu and AHT are a method employed, along with process of elimination removing each hardware component to try "identify" the fault, often little or no success has been acquired. Suggestions to clean each and every connector is good housekeeping, removing dust and build up enables better connectivity and provides better airflow and cooling. It must be assumed that as Apple, like other turnkey computer manufacturers source compatible parts to complete the system, it would be highly likely that there is a supply of one or more components that will become faulty, either by the environment, use, heat, other variables which is not easily identified as the fault found by the end user or Apple technicians.

I have had the joy of dealing with Apple genius bar staff, apple technicians, Apple Care support and other Apple Authorised Service Centres since 2010 who to the best of their knowledge cannot correctly identify the cause(s) of the random restarts of the Mac Pro, regardless of the model. I have had just about every component replaced, investigated, blammed, but to no avail, the random restart still occurs. I have the Mac Pro early 2008 3,1. Random restarts occurred within 12 months and have never stopped. The intermission period between the random restarts are not consistent, occur regardless whether an application is running or not, whether HDDs or SSD, regardless of the video card, whether A/B or only A riser has ram fitted. Generally a recall occurs when there is a safety issue, possible harm to humans, examples are exploding batteries in laptops! But a fault that causes random restarts would not be deemed as crucial enough to warrant a recall, however some investigation should be implemented to the Mac Pro users whom have experienced this random restart.

2 days ago apple announced (May 10th 2014) not publicly though, that they will not be providing support or parts any longer for the 2008 or earlier Mac Pro. They called it "Antique". Interestingly, although not perceived as accurate to some industries, Wikipedia describe Antique as:

An antique (Latin: antiquus; "old", "ancient") is an old collectable item. It is collected or desirable because of its age, beauty, rarity, condition, utility, personal emotional connection, and/or other unique features. It is an object that represents a previous era or time period in human society.

It is common practice to define "antique" as applying to objects at least 100 years old. However it goes on to say :

The term "antique" is pejorative in some instances to imply that something is out of date.

Some would say, why don't you just upgrade, by a new one? Thats what the nice young fellow at the Apple Store said, my response was "you pay for it". Just because something is 6 years old, but has been kept up to date operating system wise still makes it fit for current use, otherwise apple would not support it under its compatibility with mavericks, for example.

Getting back to this topic of Random Restarts for Mac Pro users, there is possibly a common cause that some nice technician has identified across a number of Mac Pros that might like to shed light.

My 2006 Honda Accord needs a new part, perhaps one should just buy a new one. There are laws in each country that govern how long a manufacturer must be able to service and supply parts for a product. If no service or parts can be provided and the product is expected to last longer than it has then the consumer has rights to a replacement or upgrade that would provide what the faulty product was providing. Not a legal statement just summarising.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.