June 29, 20251 yr Yesterday I noticed that some dockers weren't accessible, so I decided to reboot the server. It hung on mounting disk 2 all day. I ran some pool status commands and there appeared to be nothing wrong with the drive. In frustration I decided to roll back to version 6.12.15 as the array was much faster to start and stop before version 7 for some reason. This did nothing to change the system getting hung up on mounting disk 2. I then decided to go back to the latest version so I copied the files back into the root of the flash drive and tried to reboot. I started getting checksum errors from the boot files and couldn't boot so I had my son in law put the flash drive in the kvm that I have attached so I could rebuild it that way. This is where I took a very dumb wrong turn. When downgrading the system it overwrote some of the config and so I decided to use the latest flash backup I had available. Unfortunately I had a couple of drives that I changed out since that backup. After realizing my mistake I found the data recovery page that says don't use an old backup but it was too late. I decided that the best course of action was to delete the config/super.dat file and reassign the drives the way I had them before all this started. Things were looking like it would be ok as it showed that it would rebuild the 2 parity drives once the array started. Rebooted server and still getting stuck mounting disk 2 for some reason.At this point here is what I want to do and am looking for advice on how to proceed instead of just blindly running ahead and making things worse. My current setup is 8 12 tb drives set up as 2 parity and 6 data.I ordered 2 20 tb drives and want switch out 2 of the 12 tb drives with them. I am thinking 1 parity and 7 data drives.My thoughts at this point is that when I get home next week and can do the changes I will tell unraid to do a new configuration and only add the 2 20 tb drives to the array. I will then mount disk 1 and move the data from it to the 1st 20 tb disk. Once that is done add that disk into the array and repeat until I have the entire array back to where it should be. That's the only way I can think to get around disk 2 not mounting so the array can start and rebuild the parity drives.
June 29, 20251 yr Start by posting your diagnostics for the experts. Also did you run check filesystem from the GUI for disk 2?
June 30, 20251 yr Author There are no options in the gui for disk 2. It is formatted zfs. Here is output from zpool status and zpool importroot@Villastromoz:~# zpool statusno pools availableroot@Villastromoz:~# zpool import pool: cache-protected id: 7165692313386357785 state: ONLINEstatus: Some supported features are not enabled on the pool. (Note that they may be intentionally disabled if the 'compatibility' property is set.)action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'.config: cache-protected ONLINE mirror-0 ONLINE sdb1 ONLINE sdc1 ONLINE pool: disk6 id: 15280495122937660975 state: ONLINEstatus: Some supported features are not enabled on the pool. (Note that they may be intentionally disabled if the 'compatibility' property is set.)action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'.config: disk6 ONLINE md6p1 ONLINE pool: disk4 id: 5159517906433113695 state: ONLINEstatus: Some supported features are not enabled on the pool. (Note that they may be intentionally disabled if the 'compatibility' property is set.)action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'.config: disk4 ONLINE md4p1 ONLINE pool: disk3 id: 2051210034590522200 state: ONLINEstatus: Some supported features are not enabled on the pool. (Note that they may be intentionally disabled if the 'compatibility' property is set.)action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'.config: disk3 ONLINE md3p1 ONLINE pool: disk2 id: 10557574453833155549 state: ONLINEstatus: Some supported features are not enabled on the pool. (Note that they may be intentionally disabled if the 'compatibility' property is set.)action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'.config: disk2 ONLINE md2p1 ONLINE pool: disk1 id: 3506282359822825917 state: ONLINEstatus: Some supported features are not enabled on the pool. (Note that they may be intentionally disabled if the 'compatibility' property is set.)action: The pool can be imported using its name or numeric identifier, though some features will not be available without an explicit 'zpool upgrade'.config: disk1 ONLINE md1p1 ONLINEHere is the log files from before and after my messing up the flash drive. The one from yesterday is before, and the one from today is after I messed with the flash drive.villastromoz-diagnostics-20250629-1820.zipvillastromoz-diagnostics-20250628-1505.zip
June 30, 20251 yr Author 4 hours ago, JorgeB said:See if the disk imports read only, type zpool import -o readonly=on disk2The system is "hung" trying to mount disk 2 when starting the array. I can start the array in maintenance mode where the disks aren't mounted. Can this command run in maintenance mode? Also when I get home a week from today I will have 2 larger drives to put in the system in place of the current parity drives. What I would like to do is switch to a single parity drive - at least for now. I would then like to switch from zfs to xfs a disk at a time. Here is how I think I can go about it but let me run it by you just to be sure.I'm thinking I can add the 2 larger drives in and assign one of them as parity and the other as data. Then I'm thinking the path forward is to remove disk 2 from the drive assignments and start the array and let the new data disk format xfs and the parity drive rebuild from the other drives that are still assigned to the array. Once that is done I can then mount the current disk 2 as an unassigned device and move the data onto the new larger drive and then proceed to one at a time move data from disk to disk reformatting each one as I go. It seems all problems I have had so far has resulted in zfs so I really want to get away from it.If you think this is a valid path forward then I think I'll just leave things alone until next week when I can implement the changes that I want to do. I would really like to do this without having to recreate all of my dockers vms and shares if possible.
June 30, 20251 yr Recommend disabling array auto start, reboot, then try manually mounting the disk in read-only mode, you can also use maintain mode, but then won't be able to mount anything else to copy the data if the read-only mount works.
June 30, 20251 yr Author 1 hour ago, JorgeB said:Recommend disabling array auto start, reboot, then try manually mounting the disk in read-only mode, you can also use maintain mode, but then won't be able to mount anything else to copy the data if the read-only mount works.Ok, rebooted server. Since the parity drives are showing as needing to be rebuilt system started without starting array. In command line ran zpool import -o readonly=on disk2 and disk mounted read only just fine and pretty instantly. Here are my thoughts, tell me if you think this is a valid path forward.Once I get home next week I will install the 2 larger drives in place of the parity drives and make one of them parity and the other one data. This is the only way I will have a place to copy data from disk 2 I will then remove disk 2 from the drive assignments temporarily and start the array so it can rebuild the parity on the new drive and format the second new empty larger drive. Once that is done I can then mount the current disk 2 and copy it over to the new drive and begin the process of converting my drives to xfs one at a time.
June 30, 20251 yr 1 hour ago, Glenj48 said:system started without starting array.Not sure what you mean, if auto-array start is disabled, it cannot start.You should be able to start the array normally after mounting disk2 manually, disk2 will show unmountable in the GUI, but the data should be under /mnt/disk2, copy it to another disk or pool, then reformat that disk.
June 30, 20251 yr Author 4 minutes ago, JorgeB said:Not sure what you mean, if auto-array start is disabled, it cannot start.when the system booted it did not autostart the array as the parity drives are considered invalid due to my previous mistakes. Thank you for your help I believe that next week when I get home I will begin the process of getting things back up and running while putting in the larger drives.
June 30, 20251 yr Author 9 minutes ago, JorgeB said:You should be able to start the array normally after mounting disk2 manually, disk2 will show unmountable in the GUI, but the data should be under /mnt/disk2, copy it to another disk or pool, then reformat that disk.I want to switch all disks from zfs to xfs. I currently don't have any free drive bays to accomplish moving data around. I now have 3 new drives coming that will be at the house when I get there next week. At this point my parity drives are invalid due to my previous mistake. This is what I want to do but I need to know if this will work the way I think it will.Can I format 1 of the new larger drives as xfs and manually mount it and one of the array drives and then copy the data onto it from the array drive. Then after that is done format the old drive from xfs to zfs and manually mount it and another of the array drives and repeat the process until all of the drives have been converted to xfs. Then set up the array with the other 2 larger drives as parity and the 6 freshly formatted and copied drives as data drives and reboot to have the new parity drives rebuild from the data drives.If that won't work I guess the next best way would be to start over and add just the 3 new drives to a fresh array and then 1 by 1 copy and add the drives into the new array.
June 30, 20251 yr Solution Yes, basically follow this:https://docs.unraid.net/unraid-os/manual/storage-management/#converting-to-a-new-file-system-type
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.