XFS File system in trouble

[Roger Willcocks]

On 15 Aug 2015, at 20:00, Eric Sandeen <sandeen at sandeen.net> wrote:

> On 8/15/15 1:48 PM, Eric Sandeen wrote:
>> On 8/15/15 7:28 AM, Roger Willcocks wrote:
>>> xfs_repair 3.2.1 runs cleanly.
>>> 
>>> xfs_repair 3.1.1 complains about a load of stuff, including:
>> 
>> I wouldn't expect v3.1.1 to work at all, because:
>> ...
> 
> I take that back, a bit:  "errors vs. no errors" was for xfs_repair -n.
> 
> a "full" repair with v3.1.4 shows only:
> 
> # repair/xfs_repair /mnt/test2/leslie/md0.img 
> Phase 1 - find and verify superblock...
> Phase 2 - using internal log
>        - zero log...
>        - scan filesystem freespace and inode maps...
> non-null user quota inode field in superblock 5
> reset bad sb for ag 5
> non-null user quota inode field in superblock 10
> reset bad sb for ag 10
> non-null user quota inode field in superblock 18
> reset bad sb for ag 18
> non-null user quota inode field in superblock 23
> reset bad sb for ag 23
>        - found root inode chunk
> Phase 3 - for each AG...
>        - scan and clear agi unlinked lists...
>        - process known inodes and perform inode discovery...
>        - agno = 0
> ...
>        - agno = 29
> bad version number 0x3 on inode 124656869424
> bad version number 0x3 on inode 124656869424, resetting version number
> 

Yes, 3.1.1 without ‘-n’ reports fewer errors; I should have mentioned that. 


> latest git does indeed find no errors; I'll re-check with full repairs to see when the detection stopped, and see if it's a false positive or not.
> 
> -Eric
>