Docker/Cache goes Read Only after 1-2 weeks uptime

[dja]

I'm having a strange issue where my dockers/cache go R/O after about a week or two uptime. I have my dockers stored exclusively on cache.  My cache drive are 2 Sabrent 2 TB NVME drives setup in a RAID 0 pool. (So, 4tb..) One of them is connected via onboard slot and another is via an addon card.  Everything works fine after stopping and starting the array or reboot. 

Should I just roll with 1 NVME cache drive or change my cache config to have redundancy?  Quick smart tests come back ok. Diags attached. 

zeus-diagnostics-20220121-1233.zip

[JorgeB]

Btrfs is detecting data corruption:

 

Jan 16 08:50:46 ZEUS kernel: BTRFS info (device nvme0n1p1): bdev /dev/nvme0n1p1 errs: wr 0, rd 0, flush 0, corrupt 52, gen 0
Jan 16 08:50:46 ZEUS kernel: BTRFS info (device nvme0n1p1): bdev /dev/nvme1n1p1 errs: wr 0, rd 0, flush 0, corrupt 56, gen 0

 

This is usually a RAM issue, start by running memtest.

[dja]

Offfff. Lots of errors. Still running. Does it offer the option to save a log once completed? I'm hoping it is just one stick. 

[JorgeB]

Once it detects one error you can stop, test one DIMM at a time.

[dja]

Well, found the problem child. (thanks!) Here's hoping Corsair will honor the warranty.  And thank god Microcenter keeps track of your purchases since I don't have the receipt! 

 

I have 4 of these installed...
https://www.microcenter.com/product/616297/Vengeance_RGB_Pro_32GB_(2_x_16GB)_DDR4-3200_PC4-25600_CL16_Dual_Channel_Desktop_Memory_Kit_CMW32GX4M2E3200C16_-_Black