January 30, 20242 yr I recently added a new drive (4TB) to my server. Parity built successfully and the drive was operational for a few days, with ~150GB on the drive so far. Today I rebooted my server in order to add another drive (8TB), but when the server can back online, the previous 4TB drive showed up as 'wrong' with a warning that all data would be deleted when I start the array. After reading some similar forum posts, I created a new config with all my existing drive assignments and started the array. All my other drives mounted and parity started rebuilding, but the 4TB drive is showing as 'Unmountable: Unsupported or no file system'. 1) Is it safe to let the parity rebuild continue? I currently have it paused until I'm certain I won't risk data loss. 2) How can I repair the file system on the 4TB drive or otherwise make it mountable? The 4TB drive is attached to one of these, which is what I suspect to be the cause of the issue in the first place. However, I did use this exact hard drive and card on my old OpenMediaVault server for years, so I doubt it's a hardware issue. unraidnas-diagnostics-20240129-2010.zip
January 30, 20242 yr Community Expert Marvell is notorious for not updating their Linux drivers, so we don't recommend them. Doesn't explain why it thought disk2 was wrong though. 15 minutes ago, luigman said: add another drive (8TB) I don't see that one. Did you remove it? Is it the reason you added the Marvel? Was the Marvel in use before it started saying disk2 was wrong? It's debatable whether you have enough disks to justify parity2. Would you have enough motherboard ports without it?
January 30, 20242 yr Author The 8TB disk never showed up, which is another reason I think it's the pcie adapter, since both new drives are connected to it. But the 4TB drive was working using the pcie adapter before I rebooted. 39 minutes ago, trurl said: It's debatable whether you have enough disks to justify parity2. Would you have enough motherboard ports without it? I'm planning on adding more disks soon, as I migrate things off my old server. Plus I'd just like to be resilient to two drive failures. I do have enough ports on the motherboard to fit all the drives (excluding the 8TB which hasn't been added to the array yet)
January 30, 20242 yr Community Expert 14 minutes ago, luigman said: I do have enough ports on the motherboard to fit all the drives Maybe we should do that before attempting other fixes. How far has parity rebuild progressed?
January 30, 20242 yr Community Expert Or we could just see what check filesystem on disk2 would do with -n (nomodify) and decide what to do next. Doesn't matter much whether that happens before or after parity build so we could do that now. Check filesystem on disk2. Use the webUI and not the command line. Post output.
January 30, 20242 yr Author Parity sync finished 10GB (0.1%) before I paused it. Should I cancel the parity sync and shutdown, then connect all drives to the motherboard?
January 30, 20242 yr Author Just now, trurl said: Check filesystem on disk2. Use the webUI and not the command line. Post output. Ok I can run this first
January 30, 20242 yr Community Expert Just now, luigman said: Should I cancel the parity sync and shutdown, then connect all drives to the motherboard? OK, then post new diagnostics
January 30, 20242 yr Community Expert Just now, luigman said: Ok I can run this first That will be fine too.
January 30, 20242 yr Author Hmm I don't see any 'Check Filesystem Status' option. I have the array in maintenance mode
January 30, 20242 yr Community Expert It doesn't know what filesystem to use. I am guessing it was XFS since that is your default. Is that correct?
January 30, 20242 yr Community Expert I always say use the webUI since it's easy to get the command line wrong, but we can do it from the command line. Post the output of xfs_repair -nv /dev/md2p1
January 30, 20242 yr Author The command is still running, but the output so far is Phase 1 - find and verify superblock... bad primary superblock - bad magic number !!! attempting to find secondary superblock... .found candidate secondary superblock... unable to verify superblock, continuing... .found candidate secondary superblock... unable to verify superblock, continuing... .found candidate secondary superblock... unable to verify superblock, continuing... ..........................................
January 30, 20242 yr Author The final output was .....Sorry, could not find valid secondary superblock Exiting now.
January 30, 20242 yr Author Sure, here are the new diagnostics. Thanks for taking a look unraidnas-diagnostics-20240130-0839.zip
January 30, 20242 yr Community Expert 12 hours ago, luigman said: Today I rebooted my server in order to add another drive (8TB), but when the server can back online, the previous 4TB drive showed up as 'wrong' with a warning that all data would be deleted when I start the array. This is likely due to a known bug, new config won't work if it's that, did you save the diags when the disk was showing as wrong to confirm?
January 30, 20242 yr Author No, unfortunately I don't have the diagnostics from before I created the new config
January 30, 20242 yr Community Expert Solution Try this, stop the array, then type: sgdisk -o -a 8 -n 1:1M:0 /dev/sdX Replace X with the correct identifier for that disk, then reboot, disk will show wrong again, do another new config and see if it mounts now.
January 31, 20242 yr Author That fixed it, thank you!! I have all the drives connected to my motherboard now and parity is rebuilding. Just to confirm, in order to add additional drives and replace the Marvell, I should pick up a HBA like this one?
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.