Cache drive keeps dying

[Gracey400]

Hi All,

 

Got a really strange issue. My SSD cache drive died a week ago (unmountable btrfs error). I replaced it with a different SSD and set the system up from new and a day later the same thing happens. I repeated this process a second time with yet another SSD and same thing again.

 

I decided to have a go at using a mechanical drive. It has stayed running but has generally been pretty unstable. I checked the logs today to find more btrfs errors.

 

Does anyone have any ideas what could be causing this. I have attached a screenshot of the errors in the logs. I have rebooted it and im just waiting for it to start throwing the errors before I export the diagnstic info, which I will upload here too.

 

Thanks in advance

tower-diagnostics-20191017-0538.zip

Edited October 17, 2019 by Gracey400
Add zip file

[jpowell8672]

Did you try replacing the cable?

If using only a single cache drive, I suggest xfs instead of btrfs.

[Gracey400]

17 minutes ago, jpowell8672 said:

Did you try replacing the cable?

If using only a single cache drive, I suggest xfs instead of btrfs.

After the second drive failure I swapped the sata cable with one of my storage drives to eliminate that issue

 

I will have a go at XFS. Thankyou

[JorgeB]

Oct 17 15:32:54 Tower kernel: BTRFS warning (device sdb1): csum failed root 5 ino 649 off 2789376 csum 0xfe6e22bf expected csum 0xfa6e22bf mirror 1

 

Checksum failed would point to a hardware problem, like bad RAM, you should run memtest.

[Gracey400]

On 10/17/2019 at 4:48 PM, johnnie.black said:

Oct 17 15:32:54 Tower kernel: BTRFS warning (device sdb1): csum failed root 5 ino 649 off 2789376 csum 0xfe6e22bf expected csum 0xfa6e22bf mirror 1

 

Checksum failed would point to a hardware problem, like bad RAM, you should run memtest.

We got a winner. RAM returned 6 errors in the first test alone. I didnt think of RAM as the system was brand new.

 

Thanks for your help