Hard drive keeps giving me grief MD5


Recommended Posts

I've had things come up with this drive, MD5, three times now. I keep ending up with wayward files, random pictures, a few videos in the Lost&Found folder the xfs_repair creates. Is it time to replace this drive? Or perhaps I have something else that might be the source of these errors? Thanks.

 

unRaid Log excerpt

Jun 25 13:31:38 rubble kernel: XFS (md5): metadata I/O error in "xfs_trans_read_buf_map" at daddr 0x28d64b418 len 32 error 117
Jun 25 13:31:38 rubble kernel: XFS (md5): xfs_imap_to_bp: xfs_trans_read_buf() returned error -117.
Jun 25 13:31:38 rubble kernel: XFS (md5): Metadata corruption detected at xfs_buf_ioend+0x4c/0x95 [xfs], xfs_inode block 0x28d64b418 xfs_inode_buf_verify
Jun 25 13:31:38 rubble kernel: XFS (md5): Unmount and run xfs_repair
Jun 25 13:31:38 rubble kernel: XFS (md5): First 128 bytes of corrupted metadata buffer:
Jun 25 13:31:38 rubble kernel: 000000000b94922e: 24 1a 9c 92 6d 85 ce 6d 6f 93 4c d2 44 fc dc 3b $...m..mo.L.D..;
Jun 25 13:31:38 rubble kernel: 0000000023b350af: 98 25 67 cf 3f 77 ea c1 c5 b7 19 f8 16 ee f9 6f .%g.?w.........o
Jun 25 13:31:38 rubble kernel: 00000000bda7ee4a: 76 48 30 f5 c9 61 87 35 a3 da 2a ee a0 d8 92 c3 vH0..a.5..*.....
Jun 25 13:31:38 rubble kernel: 0000000005d43f5d: cc 6c cd 9b 9b 53 a0 69 79 01 e4 94 72 d5 4f 37 .l...S.iy...r.O7
Jun 25 13:31:38 rubble kernel: 00000000a9604dcf: aa 93 9e 81 25 4c 5d dd d7 25 b1 ba 1c c7 68 6b ....%L]..%....hk
Jun 25 13:31:38 rubble kernel: 00000000912129a9: 00 b6 ab b7 f7 3e 76 31 ad 48 42 a3 ae b1 05 df .....>v1.HB.....
Jun 25 13:31:38 rubble kernel: 000000008b0cde81: de da 64 5c 81 28 13 65 0b 6f 1f 49 78 a2 3e 33 ..d\.(.e.o.Ix.>3
Jun 25 13:31:38 rubble kernel: 000000008e83606e: b4 01 36 42 53 25 cc d8 e1 93 28 7f 0a 9c db 66 ..6BS%....(....f

 

Oh good lord, stopped array, brought back up in Maintenance mode. Did the xfs_repair status, here are hastebin results

Many entries in there. Unfortunate. If I replace the drive, will that fix any of this stuff? I have a hot standby in the server ready to replace this drive.

I suspect I'll have to run this repair, THEN replace the drive. Or the errors will simply get copied to the new drive, and I'll have to replace it anyhow.

 

Any help would be terrific, thanks!

 

rubble-smart-20200625-1924 disk 5.zip rubble-diagnostics-20200625-1914.zip

Link to comment

did the xfs-repair. Lost+found is 4GB comprised of 18689 files. FML

      root@rubble:/mnt/user/lost+found# find . -type f | wc -l
      18689

 

reassigned Disk 5 to my pre-cleared hot standby. Rebuilding drive.

 

But all those files in the lost+found. There is no way I'm going to even make a dent in figuring out what they are, where they go. Isn't this exactly what the RAID is supposed to protect? Oh, my files are gone, they just jumbled beyond any recognition. Is that right?!

Edited by dkerlee
Link to comment

Unraid can't protect against filesystem corruption, if corruption keeps happening always on the same disk it might be a good idea to re-format and create a new filesystem, but you'd need to backup all data there first, if corruption is happening on multiple disks without an apparent reason there's likely some hardware problem, like bad RAM, controller,  etc.

  • Like 1
Link to comment

Thank you very much! A shame with that corruption thing. I like the idea of not having to throw away this 4 year old 8TB drive. Can you direct me to some instructions for what you mentioned there?

18 hours ago, johnnie.black said:

re-format and create a new filesystem, but you'd need to backup all data there first

Is it as easy as limetech mentions in this thread? As I have a hot standby drive option, I've already removed the old Disk 5 (sdg) drive, and the array is rebuilding new Disk 5 (sdf). Would a deeper SMART scan identify any issues with it? Finished the rebuild on to the new Disk 5 sdf. Seems to have worked w/out a flaw.

The old sdg is still being precleared. How can I figure if it's safe to use in the future?

Link to comment
4 hours ago, dkerlee said:

I like the idea of not having to throw away this 4 year old 8TB drive.

Disk looks healthy, most likely filesystem in not a disk problem. 

 

4 hours ago, dkerlee said:

Is it as easy as limetech mentions in this thread?

Yes, but first you need to backup any data there, then restore it after the format. 

Link to comment

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.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.