[KLUG Members] SCSI errors

[members@kalamazoolinux.org]

>I have an Adaptec 29160 Ultra160 SCSI controller.
>I have three external RAID units daisy chained to it.
>The one on the end has an external terminator. The other two has 
>termination disabled.
>The fartest out is ID:0, the closest is ID:2 and the middle is ID:1
>IDs 0 and 1 are the same new models, ID 2 is my older different brand.
>Whenever I write to ID:2 everything seems fine. Whenever I write to the 
>other two I get errors almost continously.
>Nov 20 10:32:43 myhost kernel: (scsi0:A:0:0): parity error detected in 
>Data-out phase. SEQADDR(0x8b) SCSIRATE(0xc2)
>Nov 20 10:32:43 myhost kernel: ^INo terminal CRC packet recevied
>Anyone seen this before? 

Things like it.

>Anyone have any ideas how to troubleshoot this?

Are you possibly mixing wide & narrow devices on the same bus? (the "old"
cabinets and the new ones).  Is parity disabled on some devices?  Are you using
active termination?  If so, who is providing the power?  Your certain
termination is disabled on all devices?  Will it work if you disable disconnect?
 What is your total cable length?  (Three raid cages on the same bus sounds iffy).