prodigal1 wrote:
> This is really puzzling me. I want to turn an older VIA503 based
> machine into a 3.8 firewall box. I have gotten this error message
> repeatedly at the very beginning of the install process. I have used
> other bootable iso disks from different OSes such as Slackware, Win98,
> Mandriva etc. which all recognize the 15gb Fujitsu HD in the box and
> proceed happily into their respective install programs.
>
> Can anyone suggest potential reasons why 3.8 isn't seeing this IDE drive?

No, but a verbose dmesg (boot -c; verbose) might.

Joachim

On Sat, 25 Mar 2006 23:27:12 +0000, jKILLSPAM.schipper wrote:
> No, but a verbose dmesg (boot -c; verbose) might.

so far I have done
boot -c
which drops me into UKC where I do
verbose, where I get
verbose enabled, and then I do,
disable ahc, because it was hanging on probing ahc, and get
ahc disabled, then I do
quit, whereupon the kernel continues to boot
probes and results roll along until I get that nice black screen which
asks whether I'm installing, upgrading, or would like a shell
I hit I for install, accept the next two defaults and then I get
No disks found
#

prodigal1 wrote:
> On Sat, 25 Mar 2006 23:27:12 +0000, jKILLSPAM.schipper wrote:
>> No, but a verbose dmesg (boot -c; verbose) might.
>
> so far I have done
> boot -c
> which drops me into UKC where I do
> verbose, where I get
> verbose enabled, and then I do,
> disable ahc, because it was hanging on probing ahc, and get
> ahc disabled, then I do
> quit, whereupon the kernel continues to boot
> probes and results roll along until I get that nice black screen which
> asks whether I'm installing, upgrading, or would like a shell
> I hit I for install, accept the next two defaults and then I get
> No disks found
> #

Well, that's a step, but where is that dmesg? What disk controller is in
the box? Are you certain it's supported?

Joachim

On Sun, 26 Mar 2006 11:01:41 +0000, jKILLSPAM.schipper wrote:

> Well, that's a step, but where is that dmesg? What disk controller is in
> the box? Are you certain it's supported?

Thanks Joachim, it is the ide controller that isn't supported (VIA
VT82C571). I've gotten so used to things just working over the years, that
I just continued to exclude the _obvious_ cause of the primary failure
symptom from my thinking.
Oh well, on to another box.

prodigal1 wrote:
> On Sun, 26 Mar 2006 11:01:41 +0000, jKILLSPAM.schipper wrote:
>
>> Well, that's a step, but where is that dmesg? What disk controller is in
>> the box? Are you certain it's supported?
>
> Thanks Joachim, it is the ide controller that isn't supported (VIA
> VT82C571). I've gotten so used to things just working over the years, that
> I just continued to exclude the _obvious_ cause of the primary failure
> symptom from my thinking.
> Oh well, on to another box.

Or on to another controller. ;-)

Joachim

jKILLSPAM.schipper@math.uu.nl wrote:
> prodigal1 wrote:
>>
>> Thanks Joachim, it is the ide controller that isn't supported (VIA
>> VT82C571). I've gotten so used to things just working over the years, that
>> I just continued to exclude the _obvious_ cause of the primary failure
>> symptom from my thinking.
>> Oh well, on to another box.
>
> Or on to another controller. ;-)

Just a guess... is that drive a MASTER drive or a SINGLE drive?

I had issues with both OpenBSD and NetBSD on IDE machines whose single
hard disk drive was configured as a master drive instead of single drive.

Same symptoms: HDD not detected... once changed a jumper in the disk drive
all worked fine. The last example I remember is with a Dell PowerEdge 350
whose slave drive was removed without touching the jumpers on the master
drive.

Cheers,
Igor.

Igor Sobrado wrote:
>
> Just a guess... is that drive a MASTER drive or a SINGLE drive?
>
> I had issues with both OpenBSD and NetBSD on IDE machines whose single
> hard disk drive was configured as a master drive instead of single drive.
>
> Same symptoms: HDD not detected... once changed a jumper in the disk drive
> all worked fine. The last example I remember is with a Dell PowerEdge 350
> whose slave drive was removed without touching the jumpers on the master
> drive.

A side note: I am aware that the VIA controller used by the original
poster is not in the hardware compatibility list, but there is nothing
wrong in checking the current HDD settings. Perhaps it will work.

Best regards,
Igor.

Igor Sobrado wrote:
> Igor Sobrado wrote:
>
>>Just a guess... is that drive a MASTER drive or a SINGLE drive?

> A side note: I am aware that the VIA controller used by the original
> poster is not in the hardware compatibility list, but there is nothing
> wrong in checking the current HDD settings. Perhaps it will work.
>
> Best regards,
> Igor.

thanks Igor
I'll give that a whack too. It has probably been setup as master, but I
didn't even look.

prodigal1 wrote:
>
> thanks Igor
> I'll give that a whack too. It has probably been setup as master, but I
> didn't even look.

Don't know if your IDE controller is supported. Perhaps it is
compatible with one of the controllers supported, perhaps it is not.
In any case, trying is something you can do. ;-)

In my humble opinion, both NetBSD and OpenBSD require a single hard
disk drive to be configured as "single" drive instead of "master".
These operating systems are a bit tricky with hardware configuration,
but it assures they will nicely support hardware once configured.

Good luck!

Igor.

"Igor Sobrado" wrote in message
news:e0jq1c$6td$1@string1.ciencias.uniovi.es...
>
> In my humble opinion, both NetBSD and OpenBSD require a single hard
> disk drive to be configured as "single" drive instead of "master".
> These operating systems are a bit tricky with hardware configuration,
> but it assures they will nicely support hardware once configured.

Interesting. I had always thought that correctly setting up the hardware was
a prerequisite for *any* OS.

I *had* thought that "single" vs "master" was down to the IDE bus controller
and the drives, and was entirely decoupled from the OS/drivers... but I
confess I don't know for sure. (With my electronics design hat on, it would
seem pretty lame if it wasn't decoupled.)

More pragmatically, I've not noticed any difference between OSs in terms of
dealing with incorrectly-configured drives. However I'm not sure I've ever
had any...

Steve
http://www.fivetrees.com

On Fri, 31 Mar 2006 19:48:28 +0200, Igor Sobrado wrote:

> prodigal1 wrote:
>>
>> thanks Igor
>> I'll give that a whack too. It has probably been setup as master, but I
>> didn't even look.
>
> Don't know if your IDE controller is supported. Perhaps it is compatible
> with one of the controllers supported, perhaps it is not. In any case,
> trying is something you can do. ;-)
>
> In my humble opinion, both NetBSD and OpenBSD require a single hard disk
> drive to be configured as "single" drive instead of "master". These
> operating systems are a bit tricky with hardware configuration, but it
> assures they will nicely support hardware once configured.
>
> Good luck!
>
> Igor.
Son of a gun! I switched the jumper from master to cable select and
voila! I'm slicing the HD as I type. Thanks for the clue
Igor. For the record, the box is a Compaq DeskPro EN, small form factor,
P3-800. It's small and runs very quietly. It should make a more than
adequate firewall/NAT box.
Cheers

prodigal1 wrote:
> Son of a gun! I switched the jumper from master to cable select and
> voila! I'm slicing the HD as I type. Thanks for the clue

Fine!!!

> Igor. For the record, the box is a Compaq DeskPro EN, small form factor,
> P3-800. It's small and runs very quietly. It should make a more than
> adequate firewall/NAT box.

My Dell PowerEdge 350 is a P3-800 too. I like that machine as its
maximum power requirements are 75W. Current processors usually
require something between 150W and 200W... and only for the
processors themselves!!!

Your machine seems great, small and quiet. And its power requirements
should be low (it is a P3). It can be not only a firewall/NAT,
but also a small server if you want it.

Best regards,
Igor.

Steve at fivetrees wrote:
>
> Interesting. I had always thought that correctly setting up the hardware was
> a prerequisite for *any* OS.

It should, but I never had these single/master issues with Solaris x86.
Both NetBSD and OpenBSD are more tricky about how hardware has been
set up; I really appreciate this behaviour as it helps fixing some
configuration problems that, in other case, will remain undetected.

> I *had* thought that "single" vs "master" was down to the IDE bus controller
> and the drives, and was entirely decoupled from the OS/drivers... but I
> confess I don't know for sure. (With my electronics design hat on, it would
> seem pretty lame if it wasn't decoupled.)

Me too, but it certainly seems that single and master [with a single drive]
are not the same. I learned it some years ago...

> More pragmatically, I've not noticed any difference between OSs in terms of
> dealing with incorrectly-configured drives. However I'm not sure I've ever
> had any...

Solaris and Linux do not care about single/master differences, though.
But at least NetBSD and OpenBSD will do. Even if I want to install
another OS (on PCs, usually Solaris) I boot with the installation media
of one of these BSDs to isolate these problems.

Cheers,
Igor.

Steve at fivetrees wrote:
>
> Interesting. I had always thought that correctly setting up the hardware was
> a prerequisite for *any* OS.

To show that correct hardware configuration is not a prerequisite,
you can consider that the Compaq computer of prodigal1 (a P-III)
was probably running with the wrong HDD settings for years.

Igor.