Selfhosted

If it was me, I'd just go without parity temporarily and grab another drive for that when I could. A new system should be safe enough for a while, just not forever.

With that, the best you can do is glue the 1 and 2 TB drives together as a 3TB drive, and match it with an equal size partition on the 10 TB HDD. Those will have full redundancy, but not the remaining 7TB.

But you can at least have 3TB redundant, and 7TB of more risky storage. It can be used for things you can recover a different way, like game libraries, movie libraries, maybe backups of the 3TB since RAID doesn't protect against accidental deletions and modifications.

250GB cache

1 TB Pool

2 TB Pool

(2TB Parity partition + 8TB non-redundant storage partition) = 10 TB HDD

You’ll have a total of 3 TB that’s protected, and 8 TB that’s not. When you get another 10 TB drive in the future, then you’ll have a total of 13 TB that’s protected.

How I'd go above this is dependent on how much storage you expect to be using mid term/until you want/can buy another drive.

Must have 7TB ? Swap the 10tb for 2x4TB, then do 4TB parity 4+2+1TB as Data drives.

Is 3TB enough for the time being? Keep the 10TB and use as parity, 1+2TB as Data drives. When full, go for up to another 10TB as Data.

That second option is more upgradable in the future.

I'm guessing everyone meant Data drives by saying "pool". In unraid, Data drives are the ones protected by parity, in the array. Pools are "out of the array", not protected by parity.

Now your next drive can be 10TB! Not a mistake, you were just unintentionally future proofing

250GB - cache
1TB - Pool
2TB - Pool
10TB - Parity

With only 3TB in the pool you're going to be increasing the size of the pool potentially quite quickly depending on if you're ripping DVDs or Blu-Rays to it.

You'll soon need another drive to boost that pool size and I promise, that parity will keep those tears at bay if something happens.

You can set this sort of "redundancy with different-size drives without wasting a bunch of space" thing up at the block device level -- I understand that Synology's "Hybrid RAID" is a Linux system doing that. But you've got to be careful doing that; configure it wrong and you won't have redundancy.

I don't know, somewhat-surprisingly, of a software package that aims to manage a collection of disks to do this configuration at a high level, slicing up a collection of drives into smaller block devices and adding and removing disks and migrating data and such while providing guarantees that data has at least N drive redundancy.

That being said, even in such a configuration, you can only do so much. If you have one drive that's 10TB and the rest of all your drives intended for redundant storage sum to 3TB -- which is what you have -- you can't have a configuration that can handle failure of your 10TB drive and can store more than 3TB of data with redundancy, no matter how you slice and dice things. You're going to have to waste 7TB of space or store data without redundancy.

What's best to do is probably going to depend on how much you want to spend.

ZFS has a "copies=N" setting, but documentation and discussion I can find say there's no guarantee that the copies will end up on different devices (vdevs in ZFS parlance)

Temporarily use the 10tb drive as unassigned drive or another cache (in the latest version you can have multiple caches, they're not called cache anymore) until you buy another two 10tb drives. One to add as parity, the other to add to the pool, then copy back the stuff, add the original 10tb back