I followed the steps in the docs
https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/array-configuration/#removing-disks

However, I want to clear my dual 16TB drives regardless of CLI closing.

Should I use:

nohup dd bs=1M if=/dev/zero of=/dev/mdXp1 status=progress

Edited September 16, 2025Sep 16 by NathanR
edit title

You can install the Tmux plugin.

Thanks, seems like Tmux was already installed.

Any idea why progress is so slow?

I did this to not risk loosing parity. Though I suppose with two parity drives I can pull remove one disk at a time and keep parity regardless.

That's too slow, post the diags to see if there's anything logged there.

storinator-diagnostics-20250915-1228.zip

Thanks for taking a looksee

There are some writes going to disk2, don't appear enough to cause such a slowdown, but stop all writes temporarily and wait a few minutes to see if the speed increases.

I shutdown any dockers using the array.

Not sure what else would be writing to any array disks.
When I turn off the zero write (tried yesterday) the write speed doesn't change and array goes to idle.

Whole point was to not write parity. It looks like it's re-writing parity anyways (just super slowly). Maybe I just cancel and remove the disk and move on, hah.

Welp, tried stopping the array. It's stuck doing whatever it is doing.

Should I remove the temp drive1?

umount /mnt/disk1

rm /tmp/xmini.img

Edited September 15, 2025Sep 15 by NathanR

Welp. Array finally stopped.

...and I decided to rip the band aid off.

Tools > New Config > [I should have preserved the cache assignment] apply.

After re-arranging my drives to make my OCD happy.
Started array in maintenance mode, it found the encrypted xfs/zfs drives.
Then stop & start in normal mode.
It is now doing a parity-sync on both parity drives.

12 hours ago, NathanR said:

Whole point was to not write parity.

It has to write to parity when zeroing out a disk, it just should not be that slow.

Ahh yes, I misunderstood/dumb moment.

Duh, it has to write parity during writing zeros to drive; It's just after that I can remove it and force a "parity is valid checkbox".

Anyways, I'm content with my setup. But that documentation and/or feature should be looked at.
I can do it on my second server if it helps troubleshoot this issue (does anyone want to look at this further?).

Thanks!

1 hour ago, NathanR said:

But that documentation and/or feature should be looked at.

The documentation is correct, just did it myself recently.

So you cleared a 16TB disk for ~20hrs without the CLI terminal closing?

I would think a disclaimer stating that the dd command doesn't run in background would be useful.

I would wonder why mine didn't clear at full speed.

You can't imagine any improvements to the documentation at all?

Edited September 18, 2025Sep 18 by NathanR

Ohh, I see what you mean, suggest using tmux or something similar to run the dd command. That is a good idea, I understood you were saying the instructions were incorrect.

I've created a PR to include the recommendation to use TTM to run the dd command when clearing a diks, please note that anyone can contribute to the docs, so if you find anything that can be improved, please do:

GitHub

GitHub - unraid/docs: Unraid Documentation

Unraid Documentation. Contribute to unraid/docs development by creating an account on GitHub.