March 5, 20251 yr Do I have drive errors, disk errors, or do I need to fix a BTFS file system on the drive I am trying to copy data off of?
March 5, 20251 yr Author Here are the diagnostics. Lots of strange stuff going on. Thank you for your help! tower-diagnostics-20250305-0123.zip
March 5, 20251 yr Community Expert Solution You need to check filesystem on disks 1 and 7, but since btrfs is detecting data corruption on the pool, I would recommend running memtest first.
March 5, 20251 yr Author Thanks JorgeB! Isn't it odd that btrfs is detecting data corruption on the pool, when all my pool hard drives are xfs?
March 6, 20251 yr Author 18 hours ago, JorgeB said: The array disks are XFS, the pool is BTRFS. Oh yeah, thanks for pointing that out. Oddly enough, I have two btrfs cache drives and one xfs. I think I will convert them all to xfs in the future, once I get everything stable again. I really wanted to create a ZFS pool of cache drives since I have 64GB of memory, but since it's not Error Correcting, I don't want to chance the corruption.
March 6, 20251 yr Community Expert 4 minutes ago, PoppaJohn said: I really wanted to create a ZFS pool of cache drives since I have 64GB of memory, but since it's not Error Correcting, I don't want to chance the corruption. Not sure I understand this statement? Using XFS will not protect you against corruption if you have RAM issues - it just means it is more likely to go undetected.
March 12, 20251 yr Author On 3/6/2025 at 12:54 AM, itimpi said: Not sure I understand this statement? Using XFS will not protect you against corruption if you have RAM issues - it just means it is more likely to go undetected. I ran memtest 3 times and my memory passed each time. I don't think there is any problems with my memory. What I meant by the statement above was that I had read that ZFS uses a lot more RAM (for technical reasons beyond me, at the moment) and that if you have a bitflip in memory, it could get written to the pool/share. Many people said that if you don't have ECC memory, you are better off to stick with XFS.
March 12, 20251 yr Author On 3/5/2025 at 6:38 AM, JorgeB said: The array disks are XFS, the pool is BTRFS. Yes, the pool drive had a LOT of errors. I've been testing that drive for days now, and the drive appears to be fine, but I guess a couple of times that the power went out and my batteries failed in my UPS (it went off immediately.... I've replaced it now, btw) it must have corrupted the file system. The biggest problem seemed to be that the docker image was on it, and that would make the machine crash eventually. So I put a new nvme in the machine and rebuilt my docker image and put it on the new drive.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.