On Wed, Aug 20, 2008 at 11:30:20PM +0900, [EMAIL PROTECTED] wrote:
> (dropped vserver ML since the mail was always bounced)
>
> Louis Rilling:
> > The problem is still there with debian kernel 2.6.22-3-amd64 + aufs
> > 20080818.
> > Note that it disappears with kernel 2.6.25.
>
> Thank you for your report again and good info.
> Will you apply this patch which is for 2.6.22 and prints some info as
> kernel error (log level kern.err), and send me the log?
> With this patch, aufs will not stop at i_op.c:236. But it never mean the
> bug is fixed. This is just for debugging.
cd /srv/nfsroot64 succeeded, as well as further ls.
Here is the result of 'cd /srv/nfsroot64':
aufs aufs_permission:236:bash[2706]: ffff810076083ed8
aufs aufs_permission:238:bash[2706]: {ffff81007640b680, ffff810076c34490}
aufs aufs_permission:236:bash[2706]: ffff810076083ed8
aufs aufs_permission:238:bash[2706]: {ffff81007640b680, ffff810076c34490}
SysRq : Aufs
si=ffff81007bd35600
aufs: superblock
s-1: {perm 0x0, cnt 0}, aufs, flags 0x40000c00, cnt(BIAS) 0, active 1, xino 0
dev 0x14
s0: {perm 0x0, cnt 0}, ext2, flags 0x40000000, cnt(BIAS) 0, active 1, xino 1
s1: {perm 0x1, cnt 0}, ext3, flags 0x40000000, cnt(BIAS) 0, active 20, xino 1
aufs: root dentry
d-1: ///, aufs, cnt 3, flags 0x10, intent 0
i-1: i2, aufs, cnt 1, nl 23, 040755, ntfy 0, sz 1024, blk 2, ct 47616, np 0, st
0x0, f 0x0, g 0
d-1: bstart 0, bend 1, bwh -1, bdiropq -1, gen 0
d0: ///, ext2, cnt 8, flags 0x10, intent 0
i0: i2, ext2, cnt 2, nl 3, 040755, ntfy 0, sz 1024, blk 2, ct 47616, np 1, st
0x0, f 0x0, g 0
d1: //.nfsroot64-base, ext3, cnt 3, flags 0x0, intent 0
i1: i667809, ext3, cnt 2, nl 22, 040755, ntfy 0, sz 4096, blk 8, ct 23040, np
0, st 0x0, f 0x0, g 8f3e6903
aufs: isolated inode
>
> And I am guessing there are some other things I need to fix, because you
> wrote that you have some other problems...
Yes, but this is going to take way longer time to have reproducers. It currently
happens through nfs with ltp test rename14 run in parallel (16 instances) on a
4 nodes Kerrighed cluster... I'm trying to find a simpler scenario.
Thanks,
Louis
--
Dr Louis Rilling Kerlabs
Skype: louis.rilling Batiment Germanium
Phone: (+33|0) 6 80 89 08 23 80 avenue des Buttes de Coesmes
http://www.kerlabs.com/ 35700 Rennes
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
