Been having issues with (mainly) my download cache nvme drive. Also had a couple errors on my 2x cache ssds, but a scrub completes without error.

Errors are uncorrectable with a scrub, and oddly enough a smart self test just won't run, click start and it immediately stops without logging anything.

Anyone run into this before? Is it most likely the NVME failing, or bad memory?

Log file:

https://pastebin.com/icDiqAcz

TIA.

Edit: Also just noticed a ton of errors for loop2/docker img. Could this all be due to bad memory?

Edited January 5, 20242 yr by deductivemonkee

Please post the diagnostics.

My apologies, thought that I had. Here it is.

 

 

matt-server-diagnostics-20240105-1319.zip

Docker image is corrupt and needs to be recreated, since there's also data corruption on both cache devices it would be a good idea to run memtet.

Understood, thank you. This is not the first time my docker image has corrupted either, I will run memtest for sure. When recreating docker image, previous configs are saved in previous apps, and won't need to be remade from scratch, correct? Did it manually last time and would love to avoid that again.

I had a 2x16GB kit of DDR4 G.Skill memory (Got it on facebook marketplace for $50 CAD) - now I know why it was so cheap. One stick had memtest end with over 10k errors before the first pass finished, tested over 1.5hrs with only the other stick, no issues. Swap in the other stick, instant errors. Down to 16GB, but I'm glad that both weren't bad. Deleted docker image and it is currently re-installing applications.

 

Also swapped my i3-13100 with an i5-12600k that I grabbed on a boxing day sale for $200 CAD - hopefully the stock i3 intel cooler will suffice for now. Lol.

 

Thank you for the help JorgeB! Super appreciated. Will report back after docker rebuild is finished, but I'm fairly confident that the issue is resolved.