September 5, 20241 yr Hey, I had to replace my mainboard of my unraid server. The new motherboard had 2 less sata ports available, so untill I got my sata controller, I had 1 less sata drive of my sata cache pool (2x 1TB in mirror) and my Parity drive was missing. Today I installed my sata controller, connected all drives appropriately (mechanical drives on sata controller) and restarted. Now I added the sata drive back to the pool, yet now, the 2 1TB drives are seen as a 2TB cachepool. But I want them as a 1TB cachepool with redundancy. I can't find any information, and thought this would go automatically. Connecting to my server via Winscp, I still see only 1 sata drive in /mnt/ So: ``` /dev/sdc1 1.9T 371G 1.5T 20% /mnt/satacache ``` Files are not secured I believe. Ill let the parity sync run for the next 13 hours and then come back to you. But this caught my attention.
September 5, 20241 yr Author i will. But somehow the diagnostics is running sed -i on all types of files and taking forever. Some random media folder i have it is running that... Edited September 5, 20241 yr by Kees Fluitman
September 5, 20241 yr Community Expert 3 minutes ago, Kees Fluitman said: i will. But somehow the diagnostics is running sed -i on all types of files and taking forever. Some random media folder i have it is running that... Do you have mover logging enabled?
September 5, 20241 yr Author 8 minutes ago, itimpi said: Do you have mover logging enabled? sorry no. I found it. Yes was enabled. Disabled it now. Can i stop the diagnostics tool? I couldnt find it's process in htop. Im gonna force shutdown. Edited September 5, 20241 yr by Kees Fluitman
September 5, 20241 yr Community Expert 3 minutes ago, Kees Fluitman said: Can i stop the diagnostics tool? You could reboot the server (which is needed to clear the logs) and that should work I think.
September 5, 20241 yr Author 7 minutes ago, itimpi said: You could reboot the server (which is needed to clear the logs) and that should work I think. Yes i did that now. Ill send the diagnostics any moment now. What I'd actually like to do, is replace the two 1TB ssds with 2 TB ssds. Ive got them ready. But I dont want to start replacing them one after another, if there is no redundancy.
September 5, 20241 yr Author the only thing i can think of, is this: ``` diskFsProfile="single" ``` but i cant change that manually. It's greyed out. So i'd assume that needs to be multi-device mode, and probably means i'd have to move all files back to array. and then delete the cache pool and recreate it. server-diagnostics-20240905-1630.zip Edited September 5, 20241 yr by Kees Fluitman
September 5, 20241 yr Author 4 minutes ago, JorgeB said: Diags with the array started please. Oke. Here you are. I also read and found some stuff: ``` btrfs device remove /dev/sdX1 /mnt/cache ``` https://docs.unraid.net/unraid-os/manual/storage-management/#switching-the-pool-to-multi-device-mode Which could help me remove 1 drive from the pool. then connect one of the newer 2TB, then create a new multi-device pool from the 2TB, and then remove the other 1TB. edit. sorry forgot to add it. server-diagnostics-20240905-1643.zip Edited September 5, 20241 yr by Kees Fluitman
September 5, 20241 yr Community Expert 1 minute ago, Kees Fluitman said: Here you are. Diags are missing
September 5, 20241 yr Community Expert It's currently using the single profile, but if you want to used them as a mirror, you can click on the pool and then convert to raid1
September 5, 20241 yr Author 2 minutes ago, JorgeB said: It's currently using the single profile, but if you want to used them as a mirror, you can click on the pool and then convert to raid1 ah funny i couldnt find that. assumed the greyed out meant i couldnt convert. But it's under the balance tab. Thanks. Ill remove a drive afterwards and install the 2TB drive. Hope that will work in raid1 and will recognize the full 2TB after both 2TB are in. Edited September 5, 20241 yr by Kees Fluitman
September 6, 20241 yr Author 21 hours ago, JorgeB said: It's currently using the single profile, but if you want to used them as a mirror, you can click on the pool and then convert to raid1 Replacing a drive didnt work. Says a drive is missing. Maybe somehow it still has data spread. Or sth is still single (mbr?) So im moving all files to array first.
September 6, 20241 yr Community Expert 1 hour ago, Kees Fluitman said: Replacing a drive didnt work. That's expected since like mentioned it's using the single profile, you would need to convert to raid1 first.
September 8, 20241 yr Author On 9/6/2024 at 4:06 PM, JorgeB said: That's expected since like mentioned it's using the single profile, you would need to convert to raid1 first. i did that. Converted to raid1. It was shrinked to 1TB, yet i couldnt replace a drive. I believe some partition was still single. Now that I've completely finished replacing and everything. I had to delete my docker.img, but im still getting some kernel panics. I also noticed this btrfs_error.txt
September 8, 20241 yr Community Expert 9 hours ago, Kees Fluitman said: , but im still getting some kernel panics. I also noticed this If that is after deleting the docker image it's corrupt again, so do it once more.
September 8, 20241 yr Author 32 minutes ago, JorgeB said: If that is after deleting the docker image it's corrupt again, so do it once more. Did it three times. It was stable last 12 hours now. Ill keep an eye on the logs and otherwise try again. seems radarr causes the issue. ``` Sep 8 12:33:10 server kernel: BTRFS error (device loop2): bdev /dev/loop2 errs: wr 0, rd 0, flush 0, corrupt 1622, gen 0 Sep 8 12:33:10 server kernel: BTRFS warning (device loop2): csum failed root 918 ino 25066 off 1228800 csum 0x2d6b047d expected csum 0xa321433a mirror 1 Sep 8 12:33:10 server kernel: BTRFS error (device loop2): bdev /dev/loop2 errs: wr 0, rd 0, flush 0, corrupt 1623, gen 0 Sep 8 12:33:10 server kernel: BTRFS warning (device loop2): csum failed root 918 ino 25066 off 1228800 csum 0x2d6b047d expected csum 0xa321433a mirror 1 Sep 8 12:33:10 server kernel: BTRFS error (device loop2): bdev /dev/loop2 errs: wr 0, rd 0, flush 0, corrupt 1624, gen 0 Sep 8 12:33:10 server kernel: BTRFS warning (device loop2): csum failed root 918 ino 25066 off 1228800 csum 0x2d6b047d expected csum 0xa321433a mirror 1 Sep 8 12:33:10 server kernel: BTRFS error (device loop2): bdev /dev/loop2 errs: wr 0, rd 0, flush 0, corrupt 1625, gen 0 Sep 8 12:33:10 server kernel: BTRFS warning (device loop2): csum failed root 918 ino 25066 off 1228800 csum 0x2d6b047d expected csum 0xa321433a mirror 1 ``` Edited September 8, 20241 yr by Kees Fluitman
September 8, 20241 yr Author 2 hours ago, JorgeB said: Start by running memtest. thnx. ye i also realised. i put back in some older RAM i still had, extending the new mem. Then I took m out and it was all good again.
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.