Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Yesterday I opened my PC hub (after like, 5 years) to add more RAM and a new HDD. It was full of dust and balls of fluff, so I cleaned it a bit by blowing. I also removed the fan that was attached to the motherboard (I think it's there to cool the processor) and I cleaned it also by blowing and with a paintbrush, then put it back where it was (it had 4 weird plastic screws, it wasn't easy).

Then I added 2 x 2Gib of RAM (Kingston 1333MHz), and the new HDD (Western Digital Caviar Blue 1TB SATA 3). I couldn't find another wire like the one my first disk had, which was thicker, but there was one that was more flat, so I plugged it and it worked fine (I also plugged the wire that I think is for the data, which looks like many little wires glued together).

I'm running Ubuntu 12.04, and in the new disk I've just installed Windows 7.

Between yesterday and today, two reboots ocurred while working on Ubuntu. I haven't used Windows 7 that much so I don't know if it will happen there too.

So where can I begin to debug this?

share|improve this question
This sounds like maybe you didn't reattach that heatsink properly - if the system is getting a bit too warm it can spontaneously reboot like this. – Shinrai Oct 12 '12 at 21:10
also, check the ram settings in the BIOS and ensure they conform to the published SPD values for the RAM itself. – horatio Oct 12 '12 at 21:11
@Shinrai I installed some sensors in Ubuntu and I'm watching the temperature in the cores (though I have 4 cores, but I can only see 2 in the sensors), and it's normal I think (30-40°C). If the problem was the heatsink, should I be able to notice it with the sensors? – ChocoDeveloper Oct 12 '12 at 21:13
@ChocoDeveloper If the problem was the heatsink, it would manifest itself as the cores getting too hot, especially when under high load. You should be able to see that using the sensor program. – techturtle Oct 12 '12 at 21:33
Removing the heatsink without cleaning it and the processor, then re-apply fresh thermal compound is a bad idea and can lead to the issue you are having. – Moab Oct 12 '12 at 21:57
up vote 3 down vote accepted

I would guess that your problem is from at least one of the following:

  • At least one of the new memory modules is bad
  • The system is overheating
  • You have over-taxed your power supply

These are fairly easy to troubleshoot. For the memory, you can try running them one stick at a time and see if removing one of them fixes the problem. If it does, then the stick you took out is bad. For a more thorough and conclusive test, run MemTest86+ and make sure all the modules check out OK.

For the overheating, it seems from the comments that just appeared, that you have started checking the temps. If all that you removed was the fan (not the heat sink) then this is unlikely to be the cause, assuming the fan is still spinning OK. If you removed and replaced the heat sink, you may want to try reseating it again, especially with the addition of some thermal compound like Arctic Silver.

For random repowerings, however, my money would be on overtaxing the power supply. If you are pushing the edge of maxing the power it can handle, it will manifest itself at random times when the system is trying to do several things at once. This will result in an undervoltage to the motherboard which can cause a random reboot. Unless you know that you have a sufficiently strong PS, I would guess this one is most likely since you added new hardware, especially a new HDD which requires power. The fact that you mentioned there were no more of the same cables you used before would lead me to believe you may have a fairly lean-running PS. The best way to test this would be with a commercial power supply tester. Alternatively, you could try disconnecting the power to one or the other of the hard drives and see if it continues to reboot. Once you've disconnected one hard drive, you can run a stress test like Prime95 to see if you can force it to fail under the high energy usage.

share|improve this answer
+1, in light of reasonable temps I think PSU is a reasonable culprit. (I'm surprised that two DIMMs and a hard drive would push you over the edge, but it's certainly possible.) – Shinrai Oct 12 '12 at 22:03
If it is the PSU then I doubt it was the RAM that did it. However, the hard drive requires a direct connection to the PSU for independent power. That amount of power should be relatively minor, but if the PSU is failing or was running nearly at max capacity, then that little bit extra could be the final straw. – techturtle Oct 15 '12 at 17:47

You must log in to answer this question.

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