SCSI bios does not detect drive

Have you DISABLED the SCSI BIOS for the controller?  This must be enabled for a boot from SCSI operation.

At the display of the Adaptec copyright, type CTRL-A on your keyboard to get into Adaptec's BIOS.  Make sure that you have the BIOS enabled.

Try entering the card's BIOS as directed by jhance and disabled Bootable CD-ROM support in advanced Configuration.  Your mobo may require a BIOS update to support this card as a BBS (BIOS Boot Sequence) device.  Also see this article at Adaptec:
http://ask.adaptec.com/cgi-bin/adaptec_tic.cfg/php/enduser/std_adp.php?p_sid=Do1sxWhg&p_lva=&p_faqid=826&p_created=935426159&p_sp=cF9ncmlkc29ydD0mcF9yb3dfY250PTE1OCZwX3NlYXJjaF90ZXh0PTE5MTYwIGJvb3QmcF9zZWFyY2hfdHlwZT0zJnBfcHJvZF9sdmwxPTY2JnBfcHJvZF9sdmwyPTY3JnBfY2F0X2x2bDE9fmFueX4mcF9zb3J0X2J5PWRmbHQmcF9wYWdlPTE*&p_li=

Why can't I boot from my SCSI disk when an IDE / EIDE (ATAPI) disk is installed?

On most older motherboards the IDE controller takes priority over the SCSI controller in the boot order. On these old motherboards there is no CMOS / BIOS utility option to change the boot order from IDE first to SCSI first. Because of this, you must completely disable the IDE controller to boot to a SCSI hard drive connected via an Adaptec SCSI controller card.

Most new motherboards have an option in their CMOS / BIOS setup utility to arrange the boot order of the I/O controllers. These newer motherboards allow you to boot from a SCSI hard drive connected via an Adaptec SCSI controller card and also keep your IDE controller enabled for support of your IDE devices.

The boot order of the I/O controllers is a function of the motherboard's CMOS / BIOS setup - it is not a function of the SCSI controller's BIOS. Please check with the motherboard / computer manufacturer for a CMOS / BIOS update that includes the option to choose the boot order of the I/O controllers. If this feature is not available through a CMOS / BIOS update, then you must disable the IDE controllers on the motherboard when you booting from a SCSI hard drive. NOTE: You may boot from an IDE hard drive and still use your SCSI devices attached via an Adaptec SCSI controller.

Please see the following ASK article for more information:

How can I determine the boot order of my I/O controllers (IDE / SCSI)? (REF#990426-0022) (http://ask.adaptec.com/cgi-bin/adaptec_tic?solution&11-990426-0022&130-925168494)

-dog*

Check this link to see if you have the latest BIOS. The earliest dated revision describes your problem.

http://www.fic.com.tw/techsupport/bios/update/va-503+_bios.htm

I failed to say that I already updated to the latest bios verstion for both the card and the MB.  The FIC 503+ with this bios version does allow me to select SCSI as the first boot option.  

The scsi bios loads and I get a message "SCSI Bios installed" and the version number and everything, then a "no bootable scsi device found."  The machine boots up and I can see and use the drives fine, just can't figure out how to make one a boot drive.

I said:The machine boots up and I can see and use the drives fine,
just can't figure out how to make one a boot drive.

I meant: WINDOWS boots and I can see the drives fine...

Boot to a Windows boot disk and run fdisk. Make the partition on the SCSI drive active. This should then be your boot drive.

Verify that the SCSI HD is ID 0 or 1. I think that is important.

Regards
/RID

rid:
Its just dependent on how the SCSI card is configured.  Generally (with Adaptec cards at least, and pretty much all others) ID 0 is set to be the boot device and has the highest priority.  Higher numbered IDs have lower precedence, so if you had 3 SCSI HDs, ID'd 0, 1, and 4, 0 would be C:, 1 D:, and 4 E: (assuming no other harddrives installed).  There is a MSKB article concerning that somewhere on Technet.  You could theoretically set the card to boot from ID 6 and that would force it to be C: (I believe) with any other drives following in ID order.

Just FYI,

-dog*

Thanks :)
I am perhaps becoming a bit nearsighted here, having to deal mostly with older equipment. As you say, dog*, modern controller are more flexible regarding the boot device. If the controller in this case can accept any ID for a bootable device, this may be a cas of non-active partition or something more on the O/S level.

Cheers
/RID

If you boot off a WIn98 boot disk, is the drive detected? I would suspect not since that disk doesn't have a driver for the 19160 card.  Run FDISK from within Windows and you should be able to access the disk and set active partitions (you indicated that you had already created FAT32 partitions on the drive).  An active partition won't do you any good though if you don't have an OS on it.  Try this:

1. Boot to Win98.
2. Close *ALL* programs except for Explorer/Systray (esp. any Antivirus software)
3. Open a DOS prompt and type the following command, (assuming C: is your current Win98 IDE drive and D: is the partition on the SCSI card which you set as active):
xcopy c: d: /h /i /c /k /e /r /y
4. Wait for the copy to finish, depending on how many files you have and how fast your IDE drive is it might take a pretty long time.
5. Boot from a Win98 boot disk, when the boot completes you should determine what drive letter the active partition on the SCSI drive is (probably D:).  Type:
sys d:
The system will come back in a few seconds "System Transferred".
6. Reboot again, go into your BIOS.  Disable the onboard IDE ports (should be under Integrated Peripherals), then save/exit and restart.
7. Assuming everything else is set up correctly, you should now be booting to the SCSI drive.  If not, then your BIOS has some issue with booting to SCSI that the BIOS update didn't address.

This is also known as "Poor Man's Ghost" since it makes an exact replica of all your Win98 files, etc from C: to D:.

-dog*

dog,
seems like that last suggestions should work...won't get a chance to test it until later today...i'll let you know...

btw, the scsi card was set to ID # 7, so i just left it there...could that be a problem?...

I have tried the HD on both 0 and 1....

>>Verify that the SCSI HD is ID 0 or 1. I think that is important.

No, that is NOT IMPORTANT.


Verify that your SCSI BUS is PROPERLY TERMINATED.  That is REALLY IMPORTANT!!

If in doubt, describe how your SCI bus is terminated.

Clearly the bus is terminated or cmduran couldn't access the drive from within Windows without having it lockup eventually...the ID is somewhat important since cmduran is looking to boot from this drive; having the drive set to 0 will somewhat simplify configuring the SCSI BIOS to boot to this drive.  Clearly its not hard to change the ID setting for the boot device in the BIOS, but it just adheres to standards and simplifies things.

Wow jhance...did someone pee in your Cheerios this morning?

-dog*

>>Clearly the bus is terminated or cmduran couldn't access
>>the drive from within Windows without having it lockup
>>eventually

NO, NO, NO!!!  I've seen a SCSI system work ALMOST perfectly with no terminators at all.  But it would hang under certain circumstances.

Just because SCSI works in one case doesn't mean that the termination is good in general.  Perhaps the specific WORST CASE situation only happens when a certain operation occurs. Like BOOT DRIVE detection.  Who knows?

My point, don't ASSUME that the SCSI termination is correct.  

If you're having SCSI troubles, CHECK that it IS CORRECT.

Termination is provided by the self-terminating cable that Adaptec supplies with the card.  I've tried plugging the card in at the end of the cable and at the extra connector in the middle of the cable -- seems to work the same either way.

Sorry it has taken so long to get back.  I've had a lot of good suggestions but the one about termination got me thinking.  For some reason, initially I had been thinking that termination was needed outside the card and that the device was self terminating.  Figuring out how all of this was accomplished was a little involved because, as I had said earlier, the connection was through an adapter that adapted the drive's SCA-2 connection to the scsi cards 68 pin plug.  For those of you unfamiliar with a SCA-2 connection it carries both data and power through the same connector, so the adapter has the SCA-2 connector on one side and connections for the PCs power supply plus the 68 pin connector as well as jumpers pins on the other side.  So now you have a set of jumper pins on the drive and a set of jumper pins on the adapter so I had to figure out if there some jumper setting that would terminate the device.  (BTW, for anyone else that might be struggling with this configuration, the jumpers on the drive have to all be set to open for the jumpers on the adapter to be valid.)  To make a long story short, I finally figured out that the jumper setting were right and there was no setting for termination of the drive.  Hmmm, I thought.  So maybe the self terminating cable, which no had the terminator on the far side of the card, was simply backwards?  This had not occurred to me before because, like dog, I assumed that since the drive worked ok in windoze that the basic connection and termination must be right.  Not so, said jhance.  Hmmm, I thought again.  So I reversed the cable and sure enough it was as simple as that and now it works slick as a whistle.  200 pts to jhance.

jhance is the man...I guess there is a reason he's the top Topic Expert in this area (besides being a cheery, happy-go-lucky sort).

Wow...I learned something today, and its only 8:00 AM :)

-dog*

dogz..,

Thanks for the kind words...  

I wish I could tell you that I know all about SCSI problems because I'm really smart, but alas, the reason is that I've been burned more than once by improper termination.  

The darned things will often work great 99.9% of the time and will go for hours or even days without a hiccup.  

But it's that 0.01% that KILLS YOU!!!  

And they just KNOW how to crash when someone IMPORTANT is watching (like your manager or VP!).  ;-)

BTW, do you like the "new and improved" text formatting on this page?  The EE "demons" are at work again....

See:

https://www.experts-exchange.com/jsp/qManageQuestion.jsp?ta=commspt&qid=20315259

Yeah, its great...hopefully that will be fixed.  You'd think the "fixes" would be applied to the busiest TAs first, but apparently not.

-dog*