[nflug] Adaptec 2230S RAID Woes

[Robert Wolfe]

Ok, to begin with, let me begin by giving a little background.

System: http://www.supermicro.com/products/system/2U/6025/SYS-6025B-8R+V.cfm
Motherboard:  
http://www.supermicro.com/products/motherboard/Xeon1333/5000V/X7DVA-E.cfm
SCSI Controller:  
http://www.adaptec.com/en-US/products/scsi/ultra320_pcix/ASR-2230S/index.htm

These servers have 8 73GB SCSI drives in them, numbered 0 to 7 in the 
following order in the servers:

     4 5 6 7
     0 1 2 3

All drives are hot-swappable.

The OS on these servers is Debian 4.0r1 'Etch'.

Motherboard and controller BIOSes have been flashed to the latest rev level.

There are 4 RAID arrays with drives paired off in the following manner 
(reference to 'diagram' above):

0-4, 1-5, 2-6, 3-7

Now here is my dilemma.  Whenever I pull a drive from the 1-5, 2-6, and 
3-7 pairs and then reseat the drive again, that particular array 
rebuilds itself quite nicely.  However, BOTH drives in the 0-4 pair 
(which is my OS array on these particular servers), go red (ie the red 
error lights come on).  My question is is that I am not understanding 
why a drive pulled in one array should affect the drive(s) in another 
totally different array.

Any ideas anyone can give me at this point would be greatfully 
appreciated as Adaptec technical support is pretty much useless at this 
point.  Hopefully I didn't give too much detail here :)

TIA!

--
Robert Wolfe
Systems Administrator
LocalNet Corp.
CoreComm Internet Services
(517) 664-8924