January 7Jan 7 I've recently replaced two disks.The first disk failed and I got this error "Unmountable: wrong or no file system "I ended up clicking "Format will create a file system in all Unmountable disks."Lost all the data on this disk.A couple of weeks later a second disk failed. So I replaced the drive with a larger disk so I ended up selecting the option to swap the parity disk with the failed disk. This proceeded and I rebuilt the disk but now I have this issue again. I don't really want to loose the data on this disk as well.I've run unraid for 10+ years and replaced multiple disks and haven't had this issue. Is this something wrong with 7.2.3?FS: xfsExecuting file system check/sbin/xfs_repair -n '/dev/sdk1' 2>&1Phase 1 - find and verify superblock...couldn't verify primary superblock - not enough secondary superblocks with matching geometry !!!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.........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.................................found candidate secondary superblock...unable to verify superblock, continuing...Any thoughts?
January 7Jan 7 Author an 7 16:26:22 Tower emhttpd: shcmd (486): sgdisk -Z /dev/sdkJan 7 16:26:25 Tower root: Warning: The kernel is still using the old partition table.Jan 7 16:26:25 Tower root: The new table will be used at the next reboot or after youJan 7 16:26:25 Tower root: run partprobe(8) or kpartx(8)Jan 7 16:26:25 Tower root: GPT data structures destroyed! You may now partition the disk using fdisk orJan 7 16:26:25 Tower root: other utilities.Jan 7 16:26:25 Tower emhttpd: shcmd (487): sgdisk -o -a 8 -n 1:32K:0 /dev/sdkJan 7 16:26:26 Tower root: Creating new GPT entries in memory.Jan 7 16:26:26 Tower root: Warning: The kernel is still using the old partition table.Jan 7 16:26:26 Tower root: The new table will be used at the next reboot or after youJan 7 16:26:26 Tower root: run partprobe(8) or kpartx(8)Jan 7 16:26:26 Tower root: The operation has completed successfully.Jan 7 16:26:26 Tower emhttpd: shcmd (488): udevadm settleJan 7 16:26:51 Tower emhttpd: shcmd (491): xfs_admin -U generate /dev/md7p1Jan 7 16:26:51 Tower root: Metadata CRC error detected at 0x444f10, xfs_agf block 0x8bb2c89/0x200Jan 7 16:26:51 Tower root: xfs_admin: cannot init perag data (74). Continuing anyway.Jan 7 16:26:51 Tower root: Log inconsistent (didn't find previous header)Jan 7 16:26:51 Tower root: failed to find log headJan 7 16:26:51 Tower root: xlog_is_dirty: cannot find log head/tail (xlog_find_tail=5)Jan 7 16:26:51 Tower root: Jan 7 16:26:51 Tower root: ERROR: cannot find log head/tail, run xfs_repair
January 7Jan 7 Author Hi Gragorg,Thanks, I'm not overly attached to the data, important stuff is backed up just annoying. I'm more worried that it's going to keep happening every time I replace a disk.Diag dump attached. tower-diagnostics-20260107-1945.zip
January 7Jan 7 Community Expert From what I can see in the diags Disk7 was unmountable before being replaced; it then follows that the same would happen after, did you try checking filesystem before replacing it?
January 7Jan 7 Author Sorry I'm not sure. I think it was blank with 16 errors on the disk. But it was all ok a week or two earlier when I replaced the first disk.
January 7Jan 7 Community Expert If it happens again, recommend posting in the forum before replacing the disk, typically, rebuilding an unmountable disk will always result in the new disk also being unmountable.
January 7Jan 7 Author Ok thanks, think I just need to format the disk now? I'm not overly worried about the data. Just to prevent this happening again, should I run the xfs check on all the disks? Wondering if my HBA card is failing or another hardware issue. This server is pretty old. It's been running for over 10 years and it was all used when I bought it. i've been dragging my feet upgrading it, because I can't decide what I want.
January 7Jan 7 Community Expert Also for future reference, you used the wrong command to check filesystem. You must check the md device or you will invalidate parity. Always better to use the webUI instead of command line.Not entirely clear whether you could still check filesystem, even though you rebuilt an unmountable disk. The rebuild might be repairable.Check filesystem on disk7
January 7Jan 7 Author Hi Trurl,The paste I did in the first post was me trying to remove the disk from the array and I used the xfs check in unassigned devices, I pasted that because it was the same output from when I used the webgui xfs check when you click on the disk in the array.I haven't let the xfs check run for longer than 5 min, when i started this thread I had added the disk back to the array so it's currently rebuilding again, takes about 24 hours. So I will let the xfs check run for longer just to see what happens.Regards,Nick
January 7Jan 7 Community Expert Instead of rebuilding an unmountable disk, you should have done check filesystem on the emulated disk. Repairing the filesystem on the emulated disk so it is mountable means you would be rebuilding a mountable disk.
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.