[Gluster-devel] tla-907 not works for my afr configuration

[Anand Avati]

Nicolas,
  The latest TLA is undergoing some heavy modifications for replicate
performance. Please
avoid using the current repo codebase on a production system for
another week or two.

Avati

On Tue, Feb 10, 2009 at 8:35 PM, nicolas prochazka
<prochazka.nicolas at gmail.com> wrote:
> hi,
> with last tla, I've
>
> 009-02-10 15:56:31 D [afr-self-heal-data.c:983:afr_sh_data_lock] last:
> locking /Calcul.01 on subvolume brick_10.98.98.2
> 2009-02-10 15:56:31 D [afr-self-heal-data.c:983:afr_sh_data_lock] last:
> locking /Calcul.01 on subvolume brick_10.98.98.1
> 2009-02-10 15:56:31 D [afr-self-heal-data.c:935:afr_sh_data_lock_cbk] last:
> inode of /Calcul.01 on child 1 locked
> 2009-02-10 15:56:31 D [afr-self-heal-data.c:935:afr_sh_data_lock_cbk] last:
> inode of /Calcul.01 on child 0 locked
> 2009-02-10 15:56:31 D
> [afr-self-heal-common.c:170:afr_sh_print_pending_matrix] last:
> pending_matrix: [ -1 -1 ]
> 2009-02-10 15:56:31 D
> [afr-self-heal-common.c:170:afr_sh_print_pending_matrix] last:
> pending_matrix: [ -1 -1 ]
> 2009-02-10 15:56:31 E [afr-self-heal-data.c:812:afr_sh_data_fix] last:
> Unable to resolve conflicting data of /Calcul.01. Pleas
> e resolve manually by deleting the file /Calcul.01 from all but the
> preferred subvolume. Please consider 'option favorite-chil
> d <>'
> 2009-02-10 15:56:31 D [afr-self-heal-data.c:252:afr_sh_data_finish] last:
> finishing data selfheal of /Calcul.01
> 2009-02-10 15:56:31 D [afr-self-heal-data.c:228:afr_sh_data_unlock] last:
> unlocking /Calcul.01 on subvolume brick_10.98.98.2
> 2009-02-10 15:56:31 D [afr-self-heal-data.c:228:afr_sh_data_unlock] last:
> unlocking /Calcul.01 on subvolume brick_10.98.98.1
> 2009-02-10 15:56:31 D [afr-self-heal-data.c:185:afr_sh_data_unlck_cbk] last:
> inode of /Calcul.01 on child 0 locked
> 2009-02-10 15:56:31 D [afr-self-heal-data.c:185:afr_sh_data_unlck_cbk] last:
> inode of /Calcul.01 on child 1 locked
> 2009-02-10 15:56:31 D [afr-self-heal-data.c:70:afr_sh_data_done] last: self
> heal of /Calcul.01 completed
> 2009-02-10 15:56:31 D [fuse-bridge.c:368:fuse_entry_cbk] glusterfs-fuse: 34:
> LOOKUP() /Calcul.01 => 178974282 (0)
> 2009-02-10 15:56:31 D [inode.c:94:__dentry_hash] fuse/inode: dentry hashed
> VM-PERSO001-XP_SYSPREP.MR (178974282)
> 2009-02-10 15:56:31 D [inode.c:312:__inode_passivate] fuse/inode:
> passivating inode(178974282) lru=2/0 active=3 purge=0
> 2009-02-10 15:56:31 D [inode.c:293:__inode_activate] fuse/inode: activating
> inode(178974282), lru=1/0 active=4 purge=0
> 2009-02-10 15:56:31 D [fuse-bridge.c:1505:fuse_open] glusterfs-fuse: 35:
> OPEN /Calcul.01
> 2009-02-10 15:56:31 W [afr.c:605:afr_open] last: returning EIO, file has to
> be manually corrected in backend
> 2009-02-10 15:56:31 E [fuse-bridge.c:662:fuse_fd_cbk] glusterfs-fuse: 35:
> OPEN() /Calcul.01 => -1 (Input/output error)
> 2009-02-10 15:56:31 D [inode.c:312:__inode_passivate] fuse/inode:
> passivating inode(178974282) lru=2/0 active=3 purge=0
>
> however file calcul.01 is equal on two backend ( same size and same hash md5
> )
>
> Regards
> Nicolas P.
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>
>