BIOS Boot Order changes by itself

Hi,

I built a system for my daughter awhile ago using MSI's K8NGM2-FID
motherboard. Recently, she told me that, occasionally, it is booting to
a black screen.

Her system currently has 3 hard drives:

WD 80GB - IDE 0 master
Samsung 250GB - SATA
WD 500GB - SATA

She originally had a different motherboard in the machine, and just the
80GB drive, but when I rebuilt the machine with the MSI motherboard, I
added the Samsung drive, and so I installed Windows XP SP2 as the 1st
partition on the Samsung drive. I then set the boot order in the BIOS
so that the Samsung was the 1st in the boot order.

She added the WD500GB drive later.

The first time I looked into the problem, it looked like the boot order
in the BIOS had changed so that the WD 80BG drive was the 1st, and I was
able to get the machine to boot to WinXP again by just changing the boot
order back to having the WD80GB drive as the 1st.

This happened a couple of more times in the last couple of weeks, and I
flashed the BIOS from 3.50 to 3.70, hoping that that would solve the
problem of the "changing boot order", but then the problem has happened
again several times.

It seemed like whenever the boot order changed, the WD 80GB was always
set as the 1st in the boot order, so I just moved the WinXP partition to
the 1st partition of WD 80GB (My logic for this is that, for whatever
reason, the boot order was changing to some kind of default order, where
the WD 80GB drive was the 1st, so if I put WinXP on the WD 80GB, that
should always boot ok, even if the boot order got mysteriously changed
back to the "default" boot order).

We'll be watching to see if she continues to have the same problem after
that last change, but I was wondering if anyone had encountered such a
problem before, esp. with this motherboard?

I have run across a similar problem at work awhile ago, with some IBM
blades. In that case, IBM fixed the problem with a new BIOS, but I've
been searching for any info on this problem with the K8NGM2-FID, but
haven't found anything.

Thanks,
Jim

ohaya wrote:
> Hi,
>
> I built a system for my daughter awhile ago using MSI's K8NGM2-FID
> motherboard. Recently, she told me that, occasionally, it is booting to
> a black screen.
>
> Her system currently has 3 hard drives:
>
> WD 80GB - IDE 0 master
> Samsung 250GB - SATA
> WD 500GB - SATA
>
> She originally had a different motherboard in the machine, and just the
> 80GB drive, but when I rebuilt the machine with the MSI motherboard, I
> added the Samsung drive, and so I installed Windows XP SP2 as the 1st
> partition on the Samsung drive. I then set the boot order in the BIOS
> so that the Samsung was the 1st in the boot order.
>
> She added the WD500GB drive later.
>
> The first time I looked into the problem, it looked like the boot order
> in the BIOS had changed so that the WD 80BG drive was the 1st, and I was
> able to get the machine to boot to WinXP again by just changing the boot
> order back to having the WD80GB drive as the 1st.
>
> This happened a couple of more times in the last couple of weeks, and I
> flashed the BIOS from 3.50 to 3.70, hoping that that would solve the
> problem of the "changing boot order", but then the problem has happened
> again several times.
>
> It seemed like whenever the boot order changed, the WD 80GB was always
> set as the 1st in the boot order, so I just moved the WinXP partition to
> the 1st partition of WD 80GB (My logic for this is that, for whatever
> reason, the boot order was changing to some kind of default order, where
> the WD 80GB drive was the 1st, so if I put WinXP on the WD 80GB, that
> should always boot ok, even if the boot order got mysteriously changed
> back to the "default" boot order).
>
> We'll be watching to see if she continues to have the same problem after
> that last change, but I was wondering if anyone had encountered such a
> problem before, esp. with this motherboard?
>
> I have run across a similar problem at work awhile ago, with some IBM
> blades. In that case, IBM fixed the problem with a new BIOS, but I've
> been searching for any info on this problem with the K8NGM2-FID, but
> haven't found anything.
>
> Thanks,
> Jim


Hi,

BTW, I just got a response from MSI support suggesting that the problem
may be the CMOS battery being dead.

But, as I noted back to them, when it wouldn't boot, it wasn't going to
the BIOS display, where I think it would normally go and display a CMOS
checksum error message. Is that not the case with the MSI/AMI BIOS?

Jim

ohaya wrote:
>
> BTW, I just got a response from MSI support suggesting that the problem
> may be the CMOS battery being dead.
>
> But, as I noted back to them, when it wouldn't boot, it wasn't going to
> the BIOS display, where I think it would normally go and display a CMOS
> checksum error message. Is that not the case with the MSI/AMI BIOS?
>
> Jim

A weak battery can cause all kinds of odd and unrepeatable behavior.
Replacing the battery is something cheap to try. Sometimes battery
failure is caused by improper grounding. Check and see if the board is
touching the case or you have a misaligned support post.

'pider

Glasspider wrote:
> ohaya wrote:
>>
>> BTW, I just got a response from MSI support suggesting that the
>> problem may be the CMOS battery being dead.
>>
>> But, as I noted back to them, when it wouldn't boot, it wasn't going
>> to the BIOS display, where I think it would normally go and display a
>> CMOS checksum error message. Is that not the case with the MSI/AMI BIOS?
>>
>> Jim
>
> A weak battery can cause all kinds of odd and unrepeatable behavior.
> Replacing the battery is something cheap to try. Sometimes battery
> failure is caused by improper grounding. Check and see if the board is
> touching the case or you have a misaligned support post.
>
> 'pider


'pider,

I talked to my daughter (I had built the system for her), and she
reminded me that we had put in a new CMOS battery already, so we don't
think that that is the problem.

But, while talking about it, I think that I've come up with a "working
theory" about something that has been puzzling me.

We actually have 2 of these mobos: the one in her system, and also in
one of my systems, which I use daily. The problem is only occurring on
her system, so I've been trying to figure out why hers, and not mine.

So, when I was talking to her this morning, I think that I realized the
answer:

If you recall, when she was having the problem, her boot sequence was
set to boot to the Samsung 250 SATA.

On the other hand, while I also have a SATA drive in my system, my
system has always been booting off of a PATA drive.

This reinforces my "theory" that, for whatever reason, the BIOS on this
mobo is intermittently resetting to the default boot order, where the
PATA drive is booting 1st, ahead of the SATA drives.

If this is happening, it is not happening in such a way that it is
"corrupting" the BIOS settings, because we are not seeing any CMOS
checksum error on boot.

Now that I've re-configured her system so that WinXP is on the WD 80GB
PATA drive, I guess we'll see. So far, it has survived on night, and
hasn't had a problem, but I hope that this takes care of the problem for
her ...

Jim