As the title says I have a stripe pool with 4 drives total (3x 4Tb Seagate Constellation ES.3 SAS drives and a 2Tb Seagate SATA drive) that I use for some apps like Jellyfin and ARM and I store my digitalized DVD’s on. Aside from this pool I also have a separate mirror pool I use as a data NAS consisting of 2x 4Tb Seagate Constellation ES.3 SAS drives. As hardware I use my old HP Omen desktop pc (with a i7-8700, GTX1080 and a LSI 9207-8i HBA SAS controller) running TrueNAS Community 25.10.1 with the Nvidia driver patch from zzzhouuu.
I knew one of the Seagate Constellation drives in my stripe pool was a bit faulty but seemed to work fine for a bit. Now the drive starts to throw a lot of errors and shows a degraded state. I’ve tried removing this drive from the pool but the system throws and error each time stating that the drive is busy even though all apps and processes have been stopped. When trying to offline the drive first I also get an error stating there are no valid replicas.
Is there a way to remove this drive without replacing it?
You can remove a vdev, if all involved vdevs consist only of mirrors /stripes and if there is enough free space on the rest of the vdevs to receive the data from the removed vdev.
The thing is that all vdev’s are stripe and I haven’t even used 3% of the total pool size. My guess is that this is some sort of bug or the drive is to damaged to bring offline and/or remove. One option would be to just pull the drive out as I know exactly witch drive it is but I don’t know what will happen then as I don’t know if the drive even has any data on it.
There is no bug: You can NOT physically remove a drive in a stripe without replacing it or without first succesfully removing the drive logically (“Remove” from GUI).
You can NOT offline a drive in a non-redundant vdev.
If the drive is so damaged that it cannot be read properly to move its content, you’re stuck—and have basically lost the entire pool.
As stated in the first post…. When I use the remove button in the GUI for that drive I keep getting the message that the device is busy and the removing of the drive failed. That lead me to try and offline the drive which unfortunately and understandably didn’t work ether.
I guess this means my pool is done for and I will try to move the data temporarily to the NAS pool. Is there a way to preserve the dataset and ACL settings from that pool? This might save me a bunch of work.
