Page 2 of 2 FirstFirst 12
Results 11 to 14 of 14

Thread: Random crashing - how to determine the cause?

  1. #11
    Join Date
    Mar 2007
    Beans
    175
    Distro
    Ubuntu 12.04 Precise Pangolin

    Re: Random crashing - how to determine the cause?

    Quote Originally Posted by ahallubuntu View Post
    FYI, from what you said above, it's not clear whether you ran the CPU stress tester properly. It's not something you run for just one or two seconds - you should let it run for ten minutes and let the CPU get really hot. (When the CPU is running at full speed, not only will it get hot, it will also draw a lot more power than when idle.) Start the test and test status should print every few seconds; it may take longer to print the next prime number each time. That's OK; just tap the Enter key to make sure the system hasn't locked up.
    Hi again ahallubuntu, I left the CPU test for about ten to fifteen minutes the first time - what is in the screenshot above is all that appears. The cursor flashes, does not move to a new line when I press enter, and no test updates are printed on screen as you mention. This happens with all three of the CPU tests I have tried. does this mean the problem is with the cpu?

  2. #12
    Join Date
    Jul 2005
    Beans
    2,050

    Re: Random crashing - how to determine the cause?

    It could mean the CPU is overheating and locking up. Or again, it could mean the power supply is bad; a flaky power supply can cause system lockups like this, and the CPU running at full power puts a heavier load on the power supply.

    CPUs can fail but it's highly unusual especially in the scenario you describe. A CPU should either fail completely or not fail.

    It could mean the test is incompatible with your CPU, though I've never had that problem with any system I've tried to run these tests on. You've tried several tests, too. So I'd guess that isn't the issue.

  3. #13
    Join Date
    Mar 2007
    Beans
    175
    Distro
    Ubuntu 12.04 Precise Pangolin

    Re: Random crashing - how to determine the cause?

    Quote Originally Posted by ahallubuntu View Post
    It could mean the CPU is overheating and locking up. Or again, it could mean the power supply is bad; a flaky power supply can cause system lockups like this, and the CPU running at full power puts a heavier load on the power supply.

    CPUs can fail but it's highly unusual especially in the scenario you describe. A CPU should either fail completely or not fail.

    It could mean the test is incompatible with your CPU, though I've never had that problem with any system I've tried to run these tests on. You've tried several tests, too. So I'd guess that isn't the issue.
    It's an AMD CPU...nothing too unusual, I don't think. I have just updated the BIOS, and put the graphics card back in to see if this still provokes crashes. If this update hasn't settled the issue then I will present the various results to my retailer and see what they say. I think many of my components are still under warranty, so fingers crossed I can get a replacement.

    Thanks for your time and considerate replies, everyone. I'll hopefully be able to post a positive update and mark this thread solved very soon
    Last edited by Sunships; June 10th, 2012 at 10:22 PM.

  4. #14
    Join Date
    Mar 2007
    Beans
    175
    Distro
    Ubuntu 12.04 Precise Pangolin

    Re: Random crashing - how to determine the cause?

    Hello again! Updating the BIOS didn't solve any problems, and a PSU tester gadget didn't show any errors either.

    On the advice of my components retailer I took my system to my local computer repair shop, who told me that the RAM was faulty - despite it not producing errors in memtest. I had the RAM replaced twice, and this did not solve the problem - funnily enough, although I didn't see any errors with the first set of RAM (see previous posts above) both the replacement sets of RAM did give errors in memtest - LOTS of errors, like 100000s-1000000s of errors if left to run overnight.

    This in turn led me to believe that either the motherboard or processor was faulty. With Google-searching indicating the motherboard as being the much more likely candidate to fail I returned this under RMA for testing, and it passed with no faults, leaving only the CPU to send back.

    Before doing so, I thought I may as well clean off and replace the thermal paste, and see if that had any effect, since I was going to be dismantling the whole thing anyway. I had dreaded doing this, but reassembling it all was much, much more easy than when I first set up the system. Anyway, I replaced the existing paste which came pre-applied to the heat sink with some Arctic MX-3 and lo and behold, the system has not crashed since! I'm still using the same replacement RAM that gave me millions of errors too!

    I haven't run memtest again yet, but may do so just to see if I do get any errors. For now, I am just counting my blessings!

    TL, DR; suspected RAM issue turns out to be fixed by application of fresh thermal paste. Hooray!
    Last edited by Sunships; July 27th, 2012 at 01:13 PM.

Page 2 of 2 FirstFirst 12

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •