November 23, 20169 yr Hi all, having a bit of weirdness with my server....it's been humming along without incident for ages, and i rebooted it for an unraid OS update, and one of my drives came up saying "unmountable - no file system": The strange thing is it doesn't seem to be showing that the array is in a "degraded" mode (i'm used to seeing it say that it's "emulated" when a drive has gone bad). I ran SMART checks on the drive and they come back clean. I know from reading previous forum posts that formatting it is a bad idea since the parity will get rewritten to think that drive is "empty", so i haven't done that. What are the next steps?
November 23, 20169 yr Community Expert That's not a failed disk, it's file system corruption. Tools -> Diagnostics
November 23, 20169 yr Author Here's the file the diagnostics spat out. feezfileserv-diagnostics-20161123-0759.zip
November 23, 20169 yr Community Expert Check filesystem on disk3: https://lime-technology.com/wiki/index.php/Check_Disk_Filesystems#Drives_formatted_with_XFS BTW, in case your not aware, your unassigned 3TB disk has one pending sector. Device Model: WDC WD30EFRX-68EUZN0 Serial Number: WD-WCC4N2VP3HPT 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 1
November 23, 20169 yr Author Yeah, i was aware of the unassigned bit - that was part of a failed attempt to use a single unraid server to run my ESXI datastore and my fileserver "all under one roof". Even with the datastore share restricted to that drive i couldn't get the throughput i wanted, so i scrapped the idea. I'll run the filesystem check and report back. Thanks!
November 23, 20169 yr Community Expert Yeah, i was aware of the unassigned bit - that was part of a failed attempt to use a single unraid server to run my ESXI datastore and my fileserver "all under one roof". Even with the datastore share restricted to that drive i couldn't get the throughput i wanted, so i scrapped the idea. I'll run the filesystem check and report back. Thanks! A pending sector is a problem with the disk and is not related to your attempted use of the disk. Preclear it to see if you can clear the pending sector. It should be changed to a reallocated sector.
November 24, 20169 yr Author Apologies for the delay in reply - it was a 6TB drive and it took a long time for the XFS repair to run. Unfortunately, after all that running it spit out ".Sorry, could not find valid secondary superblock Exiting now." Is it possible to set the drive up as "new" and rebuild it from parity? If so, what's the best way to go about that? Also, what is the "best practice" for preventing something like this from happening in the future?
November 24, 20169 yr Community Expert Are you sure used you used the correct command? It should be: xfs_repair -v /dev/md3
November 24, 20169 yr Author Ah, i misunderstood the directions and used /dev/sda3. Will re-run with /dev/md3 and report back!
November 24, 20169 yr Author Looks like that got it sorted. Filesystem was repaired, drive mounted properly, and aside from one 0KB file, the lost+found was empty so i don't think i even lost anything. Thank you so much! As an aside, any idea what could've caused this, or what i could do to prevent it in the future?
November 24, 20169 yr Community Expert Probably due to the unRAID upgrade and the newer xfs, can detect previously undetected corruptions, the corruption itself in my experience is more commonly caused by power cuts/unclean shutdowns.
Archived
This topic is now archived and is closed to further replies.