How do I fix a Win10 Storage Space Error when in Read-Only status

Question

I set-up a storage space with 4 USB drives, sorry can't remember exactly what kind of RAID array (redundancy level) I chose. Today it disappeared from File Explorer and when I went into 'Storage Spaces' two of the drives showed a yellow Warning sign, one of which additionally said Error and something like "you should considered swapping this disk".

I have a spare USB drive so I tried to add it to the pool but it kept saying "Can't add this drive". (The drive is the same as one of the ones currently used, and brand new, so I'm pretty sure it isn't faulty). I'm short on USB ports and it seems that only the main four (used) motherboard ports are acceptable - I've previously had that same error whenever I try to connect a drive to either my additional PCI USB card or any USB hubs.

So after trying a few times, and failing to extend the pool in order to remove the problematic drives, the storage space now appears to have gone into a read-only mode - I can no longer click 'Change Settings'/ 'Manage' and I don't know what I should do:

I've seen posts about on how to set a storage pool to read-write access by using Windows PowerShell:

Get-StoragePool <PoolName> | Set-StoragePool -IsReadOnly $false'

but I don't know what my pool's name is. How/where do I find it? And what should I do after that?

I really need help on what procedure I should follow for recovery. Given that I only have four working USB ports, should I disconnect the failing drive and add the spare one into that port (without first running through the storage pool 'remove drive' option... if I manage to set it to no longer be read only.) Or is it mandatory for me to first 'Remove' the failed drive?

Any noob-friendly step-by-step guidance would be very much appreciated.

Answer 1

Ok, so let me answer my own question in case someone comes across it with a similar issue. (It's all a bit long-winded but if it's of interest, read on)

... First off, all's well that end's well. The Storage Space is back up and running, and I was right in my assumption that having it offline (with 'Warnings' next to some drives) didn't necessarily mean that I'd lost data integrity.

I also want to answer a comment that was made elsewhere were I also posted this question... why on earth am I using external USB drives and Windows10 storage pools rather than proper RAID hardware? So I've had a number of standalone NAS in the past, some are still running but all seem to eventually also die on me due and the biggest problem I've had with them is handling a back-up for them. Using an attached USB drive for a mirrored clone never reliably worked for me. I came to the conclusion that the best option would be to use a cloud back-up service. With about 25TB of data I found that services like Backblaze were just too expensive for me (you pay premium prices for a NAS drive to be covered by a business plan). However a cheap home plan covers a windows storage pool, as it's just seen by the OS as just another drive. So for my $72 a year I've currently got 25TB backed up. I'd say that's value for money. Ditto re. my use of external USB hard-drives. I've no idea why by these external drives are just so much cheaper than their bare internal counterparts. I've seen this question asked, with no conclusive answer.

Anyway, rant over. Here are my lessons learned.

1. don't disconnect the failing drive and try to just replace it with a new one. If a drive is missing then that causes the pool to go into the read-only mode shown in my screenshot. As long as all the drives are connected (even if they have Warnings) then the pool can be 'administered'. i.e. you can go into Manage and add or remove drives.

2. Read this https://docs.microsoft.com/en-us/windows-server/storage/storage-spaces/storage-spaces-states which explains the different error status of both pool and physical drives. Using windows Powershell isn't hard. It's just like a normal command line. Open it by typing Powershell in your windows search and select the 'Run as Administrator' option (on the right-hand side of the search results window) Optionally choose 'Run ISE as Administrator' this is like an IDE for powershell scripting - essentially giving you code complete, which makes it easier to enter commands.

3. Pretty much all I did type Get-PhysicalDisk to get a listing of my drives and to check their status. I noticed that one had an Operational Status of 'Lost Communication' (which was one of the drives showing with a 'Warning'). The other, the one with the Warning and Error, was showing as {Split, OK}.

4. I powered down and checked all USB connections and moved some leads to different ports. I have a Rift VR headset which uses a few ports and temporarily unplugged it. After a couple of reboots (trying different configurations) the 'Lost' drive appeared as Connected again and now with three of the four drives operational in the pool, the storage space came back online. Now with only a Warning next to one drive. NB> It turns out I'm using a Two-Way mirror. You can actually see it in my screenshot on the original post. Under the first drop down 'Storage Spaces', it says "Storage Space 24tb" (that's the 'FriendlyName' I must have given the pool when I set it up) and then below that it shows the array configuration.

5. Now to replace the failing drive. You need to click 'Prepare for removal' but ideally you want to first add a new drive. If there isn't enough storage to shift the data to the other drives this fails without giving you any error or notification (in Windows10 at least) With my 4 drives, 1 failed, two-way mirror set-up I just couldn't remove the failed drive - you'd need at least 2 and 2 (4 drives) operational, before being able to remove a fifth failing drive.

6. My problem had been that with limited USB ports, whenever I tried to add an additional drive I kept getting the following error:
   Can't prepare drives Close all applications that are accessing the drive, and then try again. The system cannot find the file specified. (0x00000002)

So yes, obviously first check that your drive is wiped (no partitions etc. which you'll see mentioned if you google the error) but for me the problem turns out to be that only some of the ports on my PC seem to work for the initial hard-drive configuration. I have 4 ports on the back of the PC, 2 on the front, and an additional 4 on a PCI USB card. The two front and the 4 on the PCI card didn't work. What I did was move some of the pool's drives to the PCI card (they still connected and worked) which then freed up the main motherboard ports on the back of the PC. I used that to connect the new drive and finally I was able to add it to the pool without the error.

7. So, after adding the drive, let the storage pool 'optimise' itself, i.e. spread out the data across all drives. I was then able to 'prepare for removal' and eventually disconnected the failing drive. I'll be adding at least one more drive for better redundancy, in order to avoid the problem I ran into where the entire pool stopped working when two of four drives had Warnings.

And I'll be keeping an eye on those warnings and swapping out drives before I get into a similar problem.

So ok, this isn't the most robust set-up but it's cost effective for me so for now I'll run with it :)

Answer 2

I was fighting with this as well.

Windows 10 Enterprise 1909 - OS Build 18363.1256 - 64 bit

I was trying to add a bunch of new drives to the storage pool (2 x 12 TB and 3 x 4 TB). The three 4TB drives added without issue but I kept getting the error Can't prepare drives - The system cannot find the file specified. 0x00000002 whenever I tried to add the two 12 TB drives.

I moved the two new 12 TB drives to a different enclosure; the same one that had the three 4 TB drives. Subsequently, that enclosure is attached to a USB port on the motherboard and not through an Anker Ultra-Slim USB 3.0 4-port Hub that the other enclosures are connected through. Once the drives were moved to that enclosure, I had no issue adding the new drives to the pool.

_{Next time I am at MS Ignite, I will have to get an explanation of this issue from someone on the Storage team.}