Derek_

Update: i suspect my fstabbing had nothing at all to do with it. It's something to do with OpenSnitch's settings. I've had to disable OpenSnitch - at least temporarily to establish the connection. Whereupon i can re-enable it. I think it's time to replace my desktop, it's gotten glitchy... too much fiddling on my part.

OK. Well i figured it out. The TLDR is:
 
Delete the pool, create a new one, only add ONE DISK, set btrfs encrypted, THEN add the second disk to the pool.
 
Here's the full procedure i used. Maybe it can be done slightly differently, but it worked for me.
 
1. With an empty cache pool (no disks assigned) - delete the pool.
2. Clear any partitions from the unassigned devices that will be used in the cache pool.
3. Create a new pool.
4. Set the pool slots to ONE disk.
5. Assign the first disk only to the pool.
6. Click on the word `cache` - choose the filesystem type - btrfs Encrypted. (password is taken from the array password automatically), click DONE.
7. There will be an open orange coloured padlock next to the word cache.
8. Change the pool slots number to 'TWO' disks.
9. Add the second disk. It will automatically acquire the values set in the first disk.
10. START the array.
11. The open orange padlock remains, and the disks are identified as "Unmountable: unsupported or no file system"
12. Scroll down - below the STOP array button there is a greyed out FORMAT button.
13. Check the disks on the left include ONLY the disks you want to include in the pool.
14. Click "Format will create a file system in all Unmountable disks" - YES I WANT TO DO THIS
15. Click OK to the warning.
16. Click Format.
 
It will take a moment to format the disks. If everything worked ok, the open orange padlock will change to a green open padlock.
 
It *should* now be good to go.
 
Could the official instructions maybe have rather more information about how to do this procedure? Or perhaps there are official instructions, somewhere else?

Taking account of the various posts (including mine), this is what i did:
 
Old Server Hardware:
Run Parity check Turn off Array auto-start Stop the VMs and Dockers, and turn off auto-start. If any hardware pass-through on the VMs, change:
a. Graphics to VNC
b. untick any `USB Devices` and `Other PCIe Devices`
c. Tools > System Devices > Untick any devices ticked for pass-through.
(On new PC, reverse the disabling changes done in 2a,b,c.)
Take screenshots/note of: Disk order (including cache disks) - note SATA ports with single parity drive doesn't matter, but the order shown in MAIN does matter (apparently the order of SATA ports used does matter if you have 2 or more parity drives - i don't have this). Network settings - also set static ip address if not already, and if multiple NICs note the correct one is selected on the new system. Backup Data and USB stick Shutdown Rename `-EFI` directory on USB stick to `EFI` on PC.  
New Server Hardware:
Memtest (this can also be done with the unRAID usb stick, but i did it before i transferred the HDDs). Review BIOS/UEFI settings:
a. Virtualisation is on, if there's IOMMU configurations, take note and adjust as appropriate (i don't have these in my new server).
b. HDD set to AHCI (before installing HDDs)
c. HDDs are detected (after installing HDDs), double-check AHCI setting.
d. Boot from USB as first/only boot device.
e. Check XHCI is enabled, per this post.
f. Left TPM as enabled.
g. Cleared SecureBoot keys, turned off SecureBoot. Boot unRAID. Check network settings (especially if connected to a monitor/keyboard) Check disks are all detected and in the correct order (including cache). Start array. Start Docker service - test each docker and turn on auto-start as desired. Start VM service. Check pass-through settings - test each VM and turn on auto-start as desired. Check access to shares/files via usual access means. Set array to auto-start if desired. Reboot to make sure everything comes back as expected. Parity check. Backup USB stick I think that's all i did. I did write notes, but as is often the case, they're a bit of a mess
 

Soooo... I figured out why DHCP wasn't working on the VLAN on the NAS only (was working elsewhere). At some point, i tagged the port `LAN`. Not `ALL`.
 
I probably did this not long after getting the router, because i couldn't figure out how to secure my VLANs while letting some traffic i wanted through, through. I got that sorted quite a while later - but i didn't remember to flip the port back to `ALL`, i forgot all about it.
 
It just popped into my head a short while ago. Tested it - all good.
 
Spent SO long on this. LOL. CRY.
 
Not sure how to mark as "solved".