My server is crashing every once in a week for a while since I have had it setup and I've been trying to figure out a solution but nothing seems to be working.

 

I have tried:

• Disabling global c-states and set Power Supply Idle Control to typical in BIOS
• Added --c6-disable to zenstates in /boot/config/go
• Disabled XMP on memory
• Disabled CPPC
• Changed docker network to ipvlan
• Running MemTest with XMP disabled for 56 hrs and got no errors.

 

Hardware:

• CPU: AMD Ryzen 7 3700x
• GPU: Nvidia RTX 2060
• MoBo: Gigabyte B450 AORUS ELITE F65a
• RAM: 2 x 8GB G.Skill Ripjaws V Memory
• Intel 660p 1TB
• 2 x Seagate Exos X16 16TB, 1 x Seagate Exos X18 16TB, 1 x old WD Black 2TB

• LSI SAS 9207-8i

• Boot: SanDisk 32GB Ultra Fit USB 3.1

chrisserver-diagnostics-20231020-2315.zip 20231001 Panic XMP Enabled.txt 20231002 Panic XMP Enabled.txt 20231009 Panic XMP Enabled.txt 20231019 Panic XMP Disabled.txt

Try with just one stick of RAM, if the same try the other one, that will basically rule out the RAM

 

 

Thanks for getting back, I have removed one of the sticks and the server crashed again after around a day of uptime. I'm trying the other stick now.

Hi, I did some more testing, I had used the other ram stick and it crashed again, also used another memory stick form another system and it also crashed these crashes happened around about of a day or two of uptime. I added back the original memory into the system and disabled the docker service and the system has been running fine for 4 days so far.

20231103 Crash.txt

You can try enabling the containers one by one and retesting.

I have upgraded my server to 6.12.6 and installed the RTL8168 Drivers and it has seemed to resolved the instability I have been experiencing. So far I have been running for 9 days with all my usual containers enabled.

I just want to revisit this, but I think have narrowed down the issue to a bad CPU. Since I have last replied in this thread, I have changed the motherboard to an MSI B550 Tomahawk and the CPU to a Ryzen 7 5700x. At first, I changed the motherboard however this didn't resolve the instability and still got CPU taint and segfault errors that I had been experiencing, I then swapped the CPU and it has seemed to fix the issue. I currently have the system running up for a month and 6 days with no taints or segfaults showing up in the server logs.