November 8, 20241 yr After upgrading to 7 beta 4 the array wont start and gives this error: "Wrong Pool State cache invalid expansion" I have a 1 TB nvme and 1 TB SSD in Pool 1 and a single 250GB in Pool 2. Ive only had unraid running a couple months, so Im not too familiar with the inner workings of it and linux. poweredge-diagnostics-20241108-1311.zip
November 8, 20241 yr Community Expert Solution The cfg for that pool is incorrect, click the pool, then "remove pool", back on main, create the pool again, assign the devices, leave the fs set to auto, start the array, and it should re-import, if it doesn't post new diags.
December 7, 20241 yr Thanks Jorge. Same happened when I upgraded to v7RC1. I was pretty concerned about data loss but your steps fixed it.
December 24, 20241 yr Thank you JorgeB. I had the same issue after upgrading to v7. Glad I found your post with the fix.
December 24, 20241 yr I'm having a similar issue. I am getting an error saying "too many wrong or missing devices" when trying to start the array. It's referencing a 24-drive 3x8 raidz3 pool I have. All of the drives are assigned correctly. If I remove the pool and recreate it with the exact same assignments, it starts fine. I do have to specify the file system and raidz3 settings though. Is this a known issue?
December 25, 20241 yr Community Expert 15 hours ago, JohnnyGrey said: I'm having a similar issue. I am getting an error saying "too many wrong or missing devices" when trying to start the array. It's referencing a 24-drive 3x8 raidz3 pool I have. All of the drives are assigned correctly. If I remove the pool and recreate it with the exact same assignments, it starts fine. I do have to specify the file system and raidz3 settings though. Is this a known issue? Assuming you are using v7, it should import the pool on auto, please post the diagnostics and the output from zpool import
December 26, 20241 yr On 11/8/2024 at 11:55 AM, JorgeB said: The cfg for that pool is incorrect, click the pool, then "remove pool", back on main, create the pool again, assign the devices, leave the fs set to auto, start the array, and it should re-import, if it doesn't post new diags. Thanks this worked for my issues of "Wrong pool state" -invalid expression. Upgrading from 6.12.10 > 7.0.0-rc.2
January 3, 20251 yr On 11/8/2024 at 12:55 PM, JorgeB said: The cfg for that pool is incorrect, click the pool, then "remove pool", back on main, create the pool again, assign the devices, leave the fs set to auto, start the array, and it should re-import, if it doesn't post new diags. Thanks This worked for me without loosing any data Thanks
January 13, 20251 yr Also had this problem. Have 4 storage pools with SSD/NVMe. Only one of them is a mix of two different brand 1GB, and that is what gave me the problem. Scared the heck out of me just removing the pool, but everything is back, and it started just fine after re-creating the pool. Thank you.
January 18, 20251 yr On 11/8/2024 at 1:55 PM, JorgeB said: The cfg for that pool is incorrect, click the pool, then "remove pool", back on main, create the pool again, assign the devices, leave the fs set to auto, start the array, and it should re-import, if it doesn't post new diags. I was stressed to have this issue after upgrading to 7.0.0 this morning. This quite easily solved the problem. Thank you.
February 2, 20251 yr I had the same error, but because my backup server does not have any cache or ZFS pool. Under Unraid 6, there was simply no cache assigned. Under Unraid 7 the array would not start. I had to explicitly set the pool setting to 0 slots. Then I could start the array. Subsequent restarts were then also fine.
February 22, 20251 yr This also helped me. Clicked 'cache', then Remove Pool. Add the pool again as Cache, and made sure the filesystem was set to 'auto'. Started the array and it is back up. Thanks!
December 6, 2025Dec 6 If anyone is still looking, I turned off the VMs and Docker. Started the array as per normal, then ran the command: btrfs balance start -dconvert=single /mnt/cache cache replace this with whatever you have your named and let it go to stand by, check the log, it will show a bunch of BTRFS messages to get an idea when it is done.Shut down the array and remove the drive. Software-wise and physically.Go to click on the good drive in that pool, then go to REMOVE pool, DO NOT erase. Apply, pool is removed, then make a new pool with the same name and add the good drive.Startup Array with VMs and Docker enabled. Edited December 6, 2025Dec 6 by HisEvilness
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.