Did I just lose a bunch of files?

[Avi_Drissman]

ReadyNAS Ultra 4+.

This morning I tried to connect to it to save a file, and found that the share was read-only. A different computer of mine stopped doing Time Machine backups at about 4am. Logs showed:

Dec  1 03:50:04 bucket kernel: EXT4-fs error (device dm-0): ext4_iget:4989: inode #52462369: block 4142678096: comm python: invalid block
Dec  1 03:50:04 bucket kernel: Aborting journal on device dm-0-8.
Dec  1 03:50:04 bucket kernel: EXT4-fs (dm-0): Remounting filesystem read-only
Dec  1 03:50:04 bucket kernel: EXT4-fs error (device dm-0): ext4_iget:4976: inode #52462371: comm python: bad extended attribute block 10303626543104
Dec  1 03:50:05 bucket kernel: EXT4-fs error (device dm-0): ext4_lookup:1043: inode #52461977: comm python: deleted inode referenced: 52462373
Dec  1 03:50:05 bucket kernel: EXT4-fs error (device dm-0): ext4_lookup:1043: inode #52461977: comm python: deleted inode referenced: 52462379
Dec  1 03:50:21 bucket kernel: EXT4-fs error (device dm-0): ext4_lookup:1043: inode #52461976: comm python: deleted inode referenced: 52462372
Dec  1 03:50:21 bucket kernel: EXT4-fs error (device dm-0): ext4_lookup:1043: inode #52461976: comm python: deleted inode referenced: 52462374
Dec  1 03:50:21 bucket kernel: EXT4-fs error (device dm-0): ext4_iget:4989: inode #52462378: block 4142682048: comm python: invalid block
Dec  1 03:50:21 bucket kernel: EXT4-fs error (device dm-0): ext4_iget:5021: inode #52462380: comm python: bogus i_mode (135300)
Dec  1 03:50:21 bucket kernel: EXT4-fs error (device dm-0): ext4_iget:4976: inode #52462382: comm python: bad extended attribute block 9655233829812

So I rebooted with a scan and got:

***** File system check forced at Mon Dec  1 09:04:42 EST 2014 *****
fsck 1.42.8 (20-Jun-2013)
e2fsck 1.42.8 (20-Jun-2013)
/dev/c/c: recovering journal
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Free blocks count wrong (1512517731, counted=1492196226).
Fix? yes

Free inodes count wrong (182203254, counted=182193565).
Fix? yes


/dev/c/c: ***** FILE SYSTEM WAS MODIFIED *****
/dev/c/c: 561763/182755328 files (2.8% non-contiguous), 1431889022/2924085248 blocks

What does it mean to lose 10,000 inodes? I see nothing in my lost+found folder.

[StephenB]

The two summary counts that were corrected are in the superblock. These were reconstructed from a full check of everything in the file system (counting up all the inodes and block allocations). There's nothing to move into lost+found, since fsck didn't delete anything. Whatever was using the inodes was already gone. https://docs.oracle.com/cd/E23823_01/ht ... 51788.html

Have you always done clean shutdowns of the NAS? Have you checked your disk health (SMART stats) recently?

If you don't have a backup, I'd suggest making one.

[Avi_Drissman]

Summary counts only? Huh. OK, I gotcha.

I kinda-sorta have a backup from when I replaced the hard drives. But I've always done clean shutdowns, and these are new hard drives. One seems to have a high "raw read error rate" but googling around seems to say that that's expected.

Thanks for the feedback. I'm going to do a compare from the backup.

[Avi_Drissman]

This got even worse. Moving forums.

viewtopic.php?f=20&t=78816