Compaq Smart Array 4200 controller question

[Andrew Heyn]

Hi,

I upgraded the firmware from 1.36 to 1.50.  In doing so, I found
out that the exact model of the array controller is
"Integrated Smart Array," not smart array 4200.

for reference,
http://h18023.www1.hp.com/support/files/server/us/download/17363.html
was the exact link I used.  I used a linux rescue disk and executed
that file from a floppy.

After upgrading the ROM, the loading of the kernel by BTX goes 5 times
faster than it did before, but I still get the same errors with the same
frequency.  I think it may be related to FreeBSD's driver, and not the
controller itself, because no problems are shown during bootup, or array
rebuilding.

I can try a kernel from a 6-CURRENT snapshot.  I'll find out if -CURRENT has
diverged uncompatably from 5.3-RELEASE's world.  If so, I can just
re-install
with a -CURRENT snapshot.

Instead of the array locking up, could it be the drives?  They're modern
drives,
but they are not the same models "approved and proven" by Compaq.

I also saw a couple ACPI related crashes, where the kernel traps while
it's displaying the trap info...One of those scrolling crashes.  I wasn't
able to save any relevent information from them, though, which I am sorry
about.
I am going to set up the machine for serial console, so this can be logged,
and I can try to reproduce the error.


Thanks,
Andrew



-----Original Message-----
From: Matthew N. Dodd [mailto:mdodd at FreeBSD.ORG]
Sent: Wednesday, February 02, 2005 8:06 AM
To: Andrew Heyn
Cc: freebsd-hardware at FreeBSD.ORG
Subject: RE: Compaq Smart Array 4200 controller question


On Wed, 2 Feb 2005, Andrew Heyn wrote:
> I was going to use that program given to me, but untarring the kernel
> source tree would cause programs to hang indefinitely on biord...and
> then I wouldnt be able to get it done.

The change to report the error was added recently but I don't see how
reporting errors could cause your problem.

I suspect the firmware on the controller is locking up.  Code in -CURRENT
will perform a register dump on command timeout, which is the usual
indicator of a lockup.

I'd go looking for the latest firmware for your controller.

--
10 40 80 C0 00 FF FF FF FF C0 00 00 00 00 10 AA AA 03 00 00 00 08 00