[clug] Rescuing old motherboards the hard way

Daniel Pittman daniel at rimspace.net
Thu Jun 24 01:05:45 MDT 2010


Paul Wayper <paulway at mabula.net> writes:

> Through a bit of handwaving and vaguery I have found a couple of machines in
> my hands. They're out of warranty, the manufacturer doesn't want them back,
> and they're perfect for an application that I have in mind for them. They
> have no VGA output but can talk through a serial console and have USB
> connectors.

Can you attach a VGA output, or a regular or USB keyboard for input?  That
might be more robust than trying to talk over the usually insanely quirky
serial console support. :/

> They boot off a pair of SATA disks RAIDed through an Adaptec board. The
> frustrating things about this are:
>  
>  1) No matter what, it seems to either want to boot off the specific
>     RAID image or locks up at the "Verify DMI Pool Data......".

That should be pretty much the last stage of BIOS before it does hand-over to
the boot loader, so ... maybe it doesn't like your boot loader?

Er, and make sure you have a partition flagged as bootable, since some BIOS
implementations hate not having one.  Maybe even try pretending it is an MSDOS
partition of some sort in the BIOS type flag?

Failing anything else, extract the MBR from the original image, disassemble
it, and see what the heck it is doing differently?  They code at that level is
pretty much straight forward.  You can even inject little BIOS interactions
easily enough if you want to try and debug it printf style.

>  2) It recognises 'TAB' to go into the console but not Ctrl-A to go 
>     into the Adaptec configuration.

I have had a few implementations that recognize ESC-A for control-A in the
serial console ... and more than a few controller extension BIOS
implementations that bypass the BIOS bits that the serial console hooks, look
directly at hardware, and just don't work with the serial console at all.

>  3) The motherboard model (NAMB-6200) is completely unrecognised 
> either on the internet or by its (most likely) manufacturer. 
>  4) None of the jumpers are documented.
>  
>  I'm convinced that there's some clever setting somewhere - a jumper, 
> maybe, or a BIOS setting - that stops me installing stuff that the
> manufacturer didn't want me to install on it.

It could be.  Try tuning the BIOS bits.  I would certainly not expect a
jumper, even if it isn't impossible.

        Daniel

-- 
✣ Daniel Pittman            ✉ daniel at rimspace.net            ☎ +61 401 155 707
               ♽ made with 100 percent post-consumer electrons


More information about the linux mailing list