On Mon, Aug 19, 2002 at 06:55:18PM -0700, Thomas Hall wrote:
> I keep getting a recurrence of a defective directory inode with the
> same xfs_repair output every time. I get this from the latest
> xfs_repair snapshot as well as three other versions thereof:
>
> Matt R Hall
> mhall@xxxxxxxxxxxxxxx
>
> PS I am not on the XFS list so please CC replies on the thread to me,
> or I'll read a web archive if there is one. If anyone suggests more
> comprehensive testing to try I'll be happy to follow through and
> figure this out.
>
This sounds to me like xfs_repair is behaving as advertised...
> Phase 2 - using internal log
> - zero log...
> - scan filesystem freespace and inode maps...
> - found root inode chunk
> - agno = 0
> - agno = 1
> - agno = 2
> - agno = 3
> - agno = 4
> - agno = 5
> - agno = 6
> - agno = 7
> - agno = 8
> - agno = 9
> - agno = 10
> - agno = 11
> - agno = 12
> - agno = 13
> - agno = 14
> - agno = 15
> - agno = 16
> - agno = 17
> - agno = 18
> - agno = 19
> - agno = 20
> - agno = 21
> - agno = 22
> - agno = 23
> - agno = 24
> - agno = 25
> - agno = 26
> - agno = 27
> - clearing existing "lost+found" inode
> - marking entry "lost+found" to be deleted
So, at this point you have a /lost+found from a previous
xfs_repair run and I assume it has several files in it...
> - agno = 0
> - agno = 1
> - agno = 2
> - agno = 3
> - agno = 4
> - agno = 5
> - agno = 6
> - agno = 7
> - agno = 8
> - agno = 9
> - agno = 10
> - agno = 11
> - agno = 12
> - agno = 13
> - agno = 14
> - agno = 15
> - agno = 16
> - agno = 17
> - agno = 18
> - agno = 19
> - agno = 20
> - agno = 21
> - agno = 22
> - agno = 23
> - agno = 24
> - agno = 25
> - agno = 26
> - agno = 27
> Phase 5 - rebuild AG headers and trees...
> - reset superblock...
> Phase 6 - check inode connectivity...
> - resetting contents of realtime bitmap and summary inodes
> - ensuring existence of lost+found directory
> - traversing filesystem starting at / ...
> rebuilding directory inode 128
Here we have to rebuild 128 (the root directory) because we removed
the old /lost+found entry and created a new one....
> - traversal finished ...
> - traversing all unattached subtrees ...
> - traversals finished ...
> - moving disconnected inodes to lost+found ...
> disconnected inode 100742368, moving to lost+found
> disconnected inode 100742370, moving to lost+found
> disconnected inode 100742387, moving to lost+found
> disconnected inode 100742390, moving to lost+found
> disconnected inode 100742407, moving to lost+found
> disconnected inode 100742420, moving to lost+found
> disconnected inode 100742422, moving to lost+found
> disconnected inode 100742429, moving to lost+found
> disconnected inode 100742432, moving to lost+found
> disconnected inode 100742464, moving to lost+found
> disconnected inode 100742482, moving to lost+found
> disconnected inode 100742484, moving to lost+found
> disconnected inode 100742486, moving to lost+found
> disconnected inode 100742490, moving to lost+found
> disconnected inode 100745856, moving to lost+found
> disconnected inode 100745903, moving to lost+found
And these are all the inodes which a previous run of xfs_repair
has stashed away in /lost+found for you to inspect and save away
somewhere more appropriate.
> done
So, if you don't remove the contents of /lost+found xfs_repair
will keep coming back with these errors.
cheers.
--
Nathan
|