F5F Stay Refreshed Hardware Desktop Random hang/crash in Ubuntu with a single RAM slot, yet Memtest86 reports no errors.

Random hang/crash in Ubuntu with a single RAM slot, yet Memtest86 reports no errors.

Random hang/crash in Ubuntu with a single RAM slot, yet Memtest86 reports no errors.

1
10th_Doctor_
Posting Freak
768
01-30-2021, 05:53 PM
#1
I recently assembled a new computer and encountered several problems during the process. Initially, I struggled to obtain any data from the screen/mobo interface to POST. After replacing the video card, it managed to connect to the BIOS and I assumed the issue was with the card itself. I then swapped it out again and installed Ubuntu 22.04 to ensure the system functioned properly and began conducting stress tests to verify all components were stable.

During these tests, I experienced increasingly frequent random crashes. The screen would abruptly turn off or go black, and the machine would become unresponsive—even when I held down the power button for over fifteen seconds. The only way to shut it down was by flipping the PSU switch or unplugging it physically, which was quite inconvenient.

Once reconnected and the PSU power restored, the system booted up again and operated normally until, at some point, it would crash again after a random interval—usually within an hour. These crashes occurred regardless of whether I was running tests or simply letting the computer sit idle. After several troubleshooting attempts and stress tests, I confirmed that storage, CPU, RAM, and PSU appear to be functioning correctly (I even used a PSU tester to verify voltages and frequencies).

This led me to suspect the motherboard might be faulty. When testing the RAM, I followed standard procedures—testing one stick at a time with PCMemTest-64 v1.5 and stress-ng. All tests passed without errors in both dual-channel configurations.

I then tried running the system with only one RAM stick in slot A2 (as recommended by the manufacturer for dual-channel use). It performed flawlessly for over six hours, passing all stress tests. Later, when I moved the single stick to slot B2, the system crashed almost immediately. This suggested a potential issue with slot B2.

I ran MemTest with both RAM sticks in slots A2 and B2, receiving no errors after eight passes. After removing the RAM from slot A2 and placing it only in slot B2, I still experienced no crashes or errors during testing.

This raises my concern: if the motherboard is defective, wouldn’t running RAM in slot B2 cause issues even when the OS is active or during MemTest? Shouldn’t there be some error message or instability? Despite multiple passes, the system remained stable. It seems possible that the problem lies with a hardware-OS incompatibility, though I haven’t encountered any such warnings.

I’m seeking advice from others who may have faced similar issues and want to know the best course of action. Should I proceed with installing Windows 11, or is it clear the motherboard is the root cause? Is MemTest able to detect any underlying problems even when RAM is in a different slot?

Thank you for any guidance you can offer.
1
10th_Doctor_
01-30-2021, 05:53 PM #1

I recently assembled a new computer and encountered several problems during the process. Initially, I struggled to obtain any data from the screen/mobo interface to POST. After replacing the video card, it managed to connect to the BIOS and I assumed the issue was with the card itself. I then swapped it out again and installed Ubuntu 22.04 to ensure the system functioned properly and began conducting stress tests to verify all components were stable.

During these tests, I experienced increasingly frequent random crashes. The screen would abruptly turn off or go black, and the machine would become unresponsive—even when I held down the power button for over fifteen seconds. The only way to shut it down was by flipping the PSU switch or unplugging it physically, which was quite inconvenient.

Once reconnected and the PSU power restored, the system booted up again and operated normally until, at some point, it would crash again after a random interval—usually within an hour. These crashes occurred regardless of whether I was running tests or simply letting the computer sit idle. After several troubleshooting attempts and stress tests, I confirmed that storage, CPU, RAM, and PSU appear to be functioning correctly (I even used a PSU tester to verify voltages and frequencies).

This led me to suspect the motherboard might be faulty. When testing the RAM, I followed standard procedures—testing one stick at a time with PCMemTest-64 v1.5 and stress-ng. All tests passed without errors in both dual-channel configurations.

I then tried running the system with only one RAM stick in slot A2 (as recommended by the manufacturer for dual-channel use). It performed flawlessly for over six hours, passing all stress tests. Later, when I moved the single stick to slot B2, the system crashed almost immediately. This suggested a potential issue with slot B2.

I ran MemTest with both RAM sticks in slots A2 and B2, receiving no errors after eight passes. After removing the RAM from slot A2 and placing it only in slot B2, I still experienced no crashes or errors during testing.

This raises my concern: if the motherboard is defective, wouldn’t running RAM in slot B2 cause issues even when the OS is active or during MemTest? Shouldn’t there be some error message or instability? Despite multiple passes, the system remained stable. It seems possible that the problem lies with a hardware-OS incompatibility, though I haven’t encountered any such warnings.

I’m seeking advice from others who may have faced similar issues and want to know the best course of action. Should I proceed with installing Windows 11, or is it clear the motherboard is the root cause? Is MemTest able to detect any underlying problems even when RAM is in a different slot?

Thank you for any guidance you can offer.

J
Juan2610
Posting Freak
875
01-30-2021, 11:56 PM
#2
Have you looked for bent pins in the CPU socket, since that could be the cause?
I’m not familiar with Ubuntu, but if it’s a hardware issue, it might not detect any problems until they occur.
J
Juan2610
01-30-2021, 11:56 PM #2

Have you looked for bent pins in the CPU socket, since that could be the cause?
I’m not familiar with Ubuntu, but if it’s a hardware issue, it might not detect any problems until they occur.