Hello all together!

I am new to this mailing list, so please be patient :-)

I own the following system:

- AMD Athlon MP 2800 + (2x -> SMP kernel)
- ASUS A7M266-D (AMD 760 MP chipset)
- 512 MB RAM Kingston KVR266X72RC25
- SCSI storage controller: Adaptec AIC-7892A U160/m
- Mass storage controller: Promise Technology, Inc. 
PDC20518/PDC40518                                  (SATAII 150 TX4)
      -> with 3 Seagate Cheetah 7200.8 250 GB attached
- ATI Technologies Inc Radeon R300 NG [FireGL X1] 256 MB
- USB Controller: NEC Corporation USB 2.0
- Ethernet controller: Realtek Semiconductor Co., Ltd. 
RTL-8139/8139C                                  /8139C+

The SATA HDDs from Seagate are working in an array, therefore see the 
following /etc/raidtab:

raiddev /dev/md0                      raid-level      5             
nr-raid-disks   3             nr-spare-disks  0             
persistent-superblock 1
       parity-algorithm        left-symmetric
       chunk-size      32            device          /dev/sdb1
       raid-disk       0             device          /dev/sdc1
       raid-disk       1
       device          /dev/sdd1
       raid-disk       2

/dev/md0 is used in a Logical Volume. pvdisplay

 PV /dev/md0   VG data   lvm2 [465.77 GB / 0    free]
 Total: 1 [465.77 GB] / in use: 1 [465.77 GB] / in no VG: 0 [0   ]

I have nearly tried everything except exchanging all the hardware :-) , 
that means changing the cables the pci-slot for the controller and all 
different kind of kernel versions: 2.6.11.12, 2.6.12 and 2.6.14.2 (the 
running kernel at this moment) and reading the SMART infos of the hdds 
(the harddisks should be ok).

When there is a high load on the RAID the following printk appears:
...
Nov 11 21:30:39 artemis ata3: status=0x51 { DriveReady SeekComplete Error }
Nov 11 21:30:39 artemis ata3: called with no error (51)!
Nov 11 21:30:39 artemis ata4: status=0x51 { DriveReady SeekComplete Error }
Nov 11 21:30:39 artemis ata4: called with no error (51)!
...

Depending on load intensity and durability there are more or less error 
messages. Especially the "called with no error (51)!" is very confusing 
for me. I have scrolled threw libata-scsi.c and it seems to me, that my 
kind of error can not be interpreted, else I shouldn't see this message, 
should I?

Perhaps you can give me any advice..... I can really need it :-)

Thanks

Michael

p.s.: The dmesg output is attached