[OpenAFS] Failed to clone the volume XXX

[Turbo Fredriksson]

I've been having trouble with my girlfriends volume(s) for a very
long time now. I always thought that it was she that did something
weird and I never managed to figure out WHY. _I_ never had any
problems (and neither did anyone else).


Two days ago, I moved my home and mail volumes around to free some
smaller disk so that they could be removed (they where just wasting
electricity :), and the night after, _I_ started having troubles
with the volumes! Looking closer, and drawing some fast conclusions
(which might be wrong, but they sound reasonable), I discovered that
I've moved my volumes to the same partition as my girlfriends volumes...

The reason it was discovered the night after, is that I get the 
following errors when doing the daily backups:

----- s n i p -----
Failed to clone the volume 536871355
Could not end transaction on the volume 536871355
Could not unlock the VLDB entry for the volume 536871355
Possible communication failure
Error in vos backup command.
Possible communication failure
----- s n i p -----

This disk is a reasonably new IBM disk (18Gb DNES-318350W if that
matters) with only one partition (sdd1) with ext3 as FS.

The file /proc/scsi/aic7xxx/1 tell me this about the disk:
----- s n i p -----
Channel A Target 10 Negotiation Settings
        User: 80.000MB/s transfers (40.000MHz, offset 255, 16bit)
        Goal: 40.000MB/s transfers (20.000MHz, offset 31, 16bit)
        Curr: 40.000MB/s transfers (20.000MHz, offset 31, 16bit)
        Channel A Target 10 Lun 0 Settings
                Commands Queued 11411628
                Commands Active 1
                Command Openings 63
                Max Tagged Openings 64
                Device Queue Frozen Count 0
----- s n i p -----

On this partition I had only some volumes that where never (or very
seldom) written to. Thinking more about this, I always got problems
with those volumes after 'some writes' to them. Don't know how MUCH
writes though since I never put the two things togheter (write = problem).


Yesterday I moved my girlfriends volumes, and she don't have any
problems any more (so far anyway :) Maybe she will, but it's very
strange that I got the same problem as she as soon as I moved
my volumes...

Both her and my volumes are very write intensive (the most write
intensive on the system) so may that have something to do with
the problem(s).

Looking through the archives again, just to be sure, I can't see
that there should be any problem with the ext3 FS...


Kernel is 2.4.21 (with XFS patch 1.3.0pre2 applied) and OpenAFS
version (including kernel module) is 1.2.6.

I know that OpenAFS is a little old, but the machine(s) live and
I refuse to fiddle with live systems :) I _could_ add (small)
patches though if this is a known error...