Intel RAID Controller ...

[Marc G. Fournier]

The server that I'm having problems with right now is an Intel server (the
motherboard and chassis combination) ... the Adaptec controller that I
have in the box, for ZCR, with Scott's latest changes, now just hangs
after a day and a bit, instead of scrolling the TIMEOUTs up the screen :(

I've tried adding the SES and aacp drivers to the kernel, to see if I can
find out what:

Apr 10 11:05:39 neptune /kernel: aac0: aac_host_command: FIB @ 0xb9ae6400
Apr 10 11:05:39 neptune /kernel: aac0:   XferState 105f<HOSTOWNED,ADAPTEROWNED,INITIALISED,EMPTY,FROMPOOL,FROMADAP,NORM>
Apr 10 11:05:40 neptune /kernel: aac0:   Command       700
Apr 10 11:05:40 neptune /kernel: aac0:   StructType    1
Apr 10 11:05:40 neptune /kernel: aac0:   Flags         0xc0
Apr 10 11:05:40 neptune /kernel: aac0:   Size          176
Apr 10 11:05:40 neptune /kernel: aac0:   SenderSize    512
Apr 10 11:05:40 neptune /kernel: aac0:   SenderAddress 0xc0631f08
Apr 10 11:05:40 neptune /kernel: aac0:   RcvrAddress   0x19400
Apr 10 11:05:40 neptune /kernel: aac0:   SenderData    0xc01d1380
Apr 10 11:05:40 neptune /kernel: aac0:    01 00 00 00 00 00 00 00 0d 00 00 00 00 00 00 00
Apr 10 11:05:40 neptune /kernel: aac0:    ff ff ff ff 06 00 00 00 d0 04 41 c0 50 05 41 c0
Apr 10 11:05:40 neptune /kernel: aac0: EventNotify(0)
Apr 10 11:05:40 neptune /kernel: aac0: (EnclosureManagement) EMPID 0 unit -1 event 6

is trying to tell me, but the aacp driver is reporting:

neptune# grep aacp /var/log/messages
Apr 10 10:58:05 neptune /kernel: aacp0: <SCSI Passthrough Bus> on aac0
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): Vendor Specific Command. CDB: 12 1 80 0 ff 0
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): ILLEGAL REQUEST asc:24,0
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): Invalid field in CDB
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): Vendor Specific Command. CDB: 12 1 80 0 ff 0
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): ILLEGAL REQUEST asc:24,0
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): Invalid field in CDB
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): Vendor Specific Command. CDB: 12 1 80 0 ff 0
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): ILLEGAL REQUEST asc:24,0
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): Invalid field in CDB
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): Vendor Specific Command. CDB: 12 1 80 0 ff 0
Apr 10 10:58:06 neptune /kernel: (probe6:aacp0:0:6:0): ILLEGAL REQUEST asc:24,0
Apr 10 10:58:07 neptune /kernel: (probe6:aacp0:0:6:0): Invalid field in CDB
Apr 10 10:58:07 neptune /kernel: (probe6:aacp0:0:6:0): Vendor Specific Command. CDB: 12 1 80 0 ff 0
Apr 10 10:58:07 neptune /kernel: (probe6:aacp0:0:6:0): ILLEGAL REQUEST asc:24,0
Apr 10 10:58:07 neptune /kernel: (probe6:aacp0:0:6:0): Invalid field in CDB
Apr 10 10:58:09 neptune /kernel: pass0 at aacp0 bus 0 target 0 lun 0
Apr 10 10:58:09 neptune /kernel: pass1 at aacp0 bus 0 target 1 lun 0
Apr 10 10:58:09 neptune /kernel: pass2 at aacp0 bus 0 target 2 lun 0
Apr 10 10:58:09 neptune /kernel: pass3 at aacp0 bus 0 target 3 lun 0
Apr 10 10:58:09 neptune /kernel: pass4 at aacp0 bus 0 target 4 lun 0
Apr 10 10:58:09 neptune /kernel: pass5 at aacp0 bus 0 target 5 lun 0
Apr 10 10:58:10 neptune /kernel: pass6 at aacp0 bus 0 target 6 lun 0
neptune# grep ses !$
grep ses /var/log/messages
neptune#

Does anyone know if the Intel ZCR Controllers have similar problems with
>2GB servers?

And, is it possible to just swap controllers?  With the RAID, I'm guessing
not, but right now I'm *really* grasping for straws here on getting this
server stable :(