H3LD3R7

Array is stopped. but it is greyed out. Cant remove the pool..

I stopped it and removed the empty drive and then started the array but it kept hanging on starting array but it never started. I had to do a reboot to get the array to stop again and there i could add the drive again. It stayed on: "Starting", mdState=STARTED. Is it safe for also here remove pool and create again? It is a shared boot/pool drive.

I bought two identical Intel DC S3500 480GB SSDs with the intention of setting up a mirrored internal boot pool with the remaining capacity as a data pool. Internal boot is working on one drive, but I have not been able to get the second drive to join. Initially the setup ran while the system was still in Legacy/CSM mode, so efibootmgr failed with rc=2 (no EFI variables). After switching the BIOS to UEFI, I ran the setup again. It completed, but reported efibootmgr failed for /dev/sdg (rc=5). zpool status pool: flash state: ONLINE NAME STATE READ WRITE CKSUM flash ONLINE 0 0 0 sde3 ONLINE 0 0 0 df -h /boot flash/boot 15G 2.6G 13G 18% /bootThe second SSD (sdg) has never been partitioned: lsblk -o NAME,SIZE,FSTYPE,LABEL /dev/sde /dev/sdg sde 447.1G ├─sde1 1M ├─sde2 510M vfat EFI ├─sde3 15.5G zfs_member flash └─sde4 431.1G sdg 447.1G Really would like the 2 drives to function properly and have the extra capacity. helder-diagnostics-20260815-1101.zip

Thanks again JorgeB helping me with this machine. Just putting it on Auto was the move.. Tried that couple of tries ago and then i also got the error that it wouldnt find the pool so from that time on i just put the pool in btrfs and it worked with 1. When i added another it just would not work. But i add screenshots it said it would format the 970 and then after i started the array it just works again! Thank you.

Thank you. I did and here are the diagnostics. Also added 2 screenshots from my main page with the issue. helder-diagnostics-20260806-1439.zip

Yes thats what i have been trying to do. Edit cache, set to 2 devices and add the 970. But it doesnt do that. I start the array and i get that screenshot i send. It will not be visible. Then i remove pool and start with the corsair and it is all good again but not with 2 drives.

So i have been having this struggle. I had a m.2 ssd intenso that was having trouble and wanting to replace it with a samsung 970 1tb. But it was a little bit smaller then the intenso 1tb.... So i tried everything with the help from claude but cant seem to get it working. I have now a running server with just my one cache drive. If i add the samsung it will show up als the samsung. What am i doing wrong? helder-diagnostics-20260806-1007.zip

It worked and i changed both drives with no issues. Thank you very much!

Just did this. Here are my diagnostics. myunraid-diagnostics-20260527-1620.zip

Ok thank you! Will try this out. Tuesday im near the machine again and then im gonna do that. Making full backup of cache now to array.

unraid:~# btrfs fi show Label: none uuid: 5e5ff1b9-820c-4d95-877f-60b4fafb86de Total devices 2 FS bytes used 209.35GiB devid 1 size 476.94GiB used 234.05GiB path /dev/sdc1 devid 2 size 465.76GiB used 228.03GiB path /dev/sdb1 Label: none uuid: b46efa3b-1e2e-4d53-a083-5161627f1866 Total devices 1 FS bytes used 54.98GiB devid 1 size 150.00GiB used 76.02GiB path /dev/loop2 Label: none uuid: 35470472-fff9-46fd-b563-4b0558caae6b Total devices 1 FS bytes used 1.14MiB devid 1 size 1.00GiB used 126.38MiB path /dev/loop3

Sorry! This is the diagnostic zip. myunraid-diagnostics-20260522-1348.zip

It keeps giving me errors when i remove one and try to start up. If i put the samsung to place 1 it also gives me the error and i coudnt just put in the new drive and boot to force it to rebuild on the new drive. What is the best way to replace the dying drive?