I’ve tried installing Manjaro a few times on my PC but for some reason, it will just randomly restart. I know it isn’t an overheating issue as I am trying to dual-boot Windows and Manjaro at the same time, and Windows seems to work perfectly fine with no overheating problems.
The timing of when the computer restarts in Manjaro seems to be pretty random, as sometimes it would let me install the OS, but when I tried updating some software it would randomly restart, and generally when I tried doing anything it would restart. Checking journalctl didn’t give me much as it only gave an error about a ‘touchegg daemon’ being unable to connect. I did at least try disabling this module by running the command ‘systemctl disable --now touchegg’ but this didn’t fix my issue.
It also doesn’t seem to be a gnome specific issue as it still happened with KDE and XFCE. I also tried installing Fedora but it still gave the same problem. I also tried both the proprietary graphics drivers and the open-source drivers just in case, but neither fixed my issue.
When the computer did restart the actual components themselves wouldn’t switch off, the screen would just randomly go blank and then go through the boot sequence again. As far as I’m aware, I think it doesn’t switch off when idle.
System specs (if helpful):
CPU: Ryzen 5 5600x
GPU: MSI RTX 3070 Gaming X Trio
Boot drive: Samsung 970 EVO 500gb
RAM: 16GB Corsair Vengeance Pro
EDIT: It was the XMP profile being turned on in the BIOS that caused the issue.
It appears to be a hardware problem rather than a software problem. I had the same problem as you and I found out with the memtest86 that I had broken RAM. Also try disabling the Precision Boost Overdrive from the bios and see if it’s a power issue.
If it happens regardless of OS, it suggests a hardware issue and not software. Quite possibly RAM, like @Dullnets said. Test your RAM with memtest86. Find out more here:
Edit:
You can also use memtester, available in the community repository:
You forgot to mention the PSU.
My opinion (as of last month) is that the PSU is the most important component - as it just took out my ancient computer - rendering the motherboard/cpu/RAM unusable (as replacements just aren’t cost efficient).
I’d seriously look at that before plugging in my 5600G/Steel Series/Samsung Evo/16GB RAM or my other hard drives (as it was the original crappy HP Pavillion power supply which crapped out my first 500GB hard disk upgrade due to insufficient power delivery).
to run memtest is a good idea. i would recommend to shut down the pc,unplug the RAM and plug it back. contact-problems with the pinnings at RAM are more common than people expect.
Hi, I did the default memtest and the results turned out fine. I found out the issue did have to do with the RAM though, as it turned out the XMP profile being turned on seemed to be causing the issue, as when I turned it off it started working fine. Thanks for the help, though!