Date: Sun, 19 Dec 2004 17:46:06 +0100
From: Philip Nienhuis <[EMAIL PROTECTED]>
Subject: Re: [LIB] When EZ-Drive is a >must< for W98 * W2K installations
Matt Hanson wrote:
>
> Date: Sat, 18 Dec 2004 16:59:14 -0800 (PST)
> From: Matt Hanson <[EMAIL PROTECTED]>
> Subject: Re: [LIB] When EZ-Drive is a >must< for W98 * W2K installations
>
> --- Matt Hanson <[EMAIL PROTECTED]> wrote:
>
> > Okay, running PM's partinfo, right off the bat I see I have a problem
> > with the Win98 OS reporting too few cylinders fot the drive. 4863
> > instead of 4864. I still have EZ-Drive installed, but I've not seen that
>
> > error in the past with EZ-D when things were working well.
> >
> > And the extended partition is type 0f.
That's needed for DOS to see beyond 8 GB. Win98/ME do not need it, they
can access 05 type extended partitions once their native disk drivers
have taken over from the DOS boot-up environment.
Anyway, 0f won't hurt - I think.
But FAT32 partitions in the extended partition beyond 8 GB need to be
type 0c ("Win95 FAT32 LBA" in Linux terms) rather than 0b (plain
"FAT32"). I've had lots of trouble when I did not obey this rule. I saw
your FAT32 partitions were 0b. I think this might be the cause of your
W98 scandisk problems.
If PM made these partitions (-types) I'd definitely dump it.
(BTW 1B = hidden FAT32)
> > Attached is the PM partinfo output file if it helps. Dan's server won't
> > forward it to the list.
>
> Just a couple of footnotes here...
>
> * The partinfo.exe for this HDD mounted in the 110 with EZD removed says
> the system reports only 1017, but shows 4864 cylinders in the partition
> table, and noted the error difference.
>
> * The partinfo.exe for this HDD mounted in the desktop with EZD removed
> says the system reports the correct full 4864 cylinders with no errors
> showing in the partition table anywhere.
>
> * Again with EZD installed and in the 110, partinfo says the system reports
> 4863 cylinders instead of the 4864 partinfo sees in the partition table,
> and makes a note of the error.
The one cylinder difference is due to a cylinder (cyl #0, the otherwise
regular MBR cylinder) which EZD reserves for itself; it shifts the rest
of the cylinders, so the system or operating systems see only 4863.
While they think the MBR is on cyl. #0, it really is on #1.
Make sure to always let EZD initialize before running any other
partitioning program or operating system. Mixing up EZD-cooked-up and
"raw" views of the HD will lead to interesting situations. Is there a
chance that such a mix-up occurred on you Lib HDD (e.g., unattended
reboot during Win2K installation)?
> For some reason the system was locking up in DOS the other night.
> Ctrl-Alt-Del didn't reboot, the power switch didn't shut it down, and tghe
> reboot switch didn't reboot. Had to pull the AC and battery pack numerous
> times.
>
> >From my experiences with both the desktop w/multiple HDDs and this 110, it
> seems a CMOS/BIOS setting thing. Sometimes the system wouldn't boot at
> all. No red 'Toshiba' splash screem. No nada. Letting the system sit,
> and I presume the CMOS to discharge, the system with finally boot again
Could it be that your Lib HW is starting to get flakey?
Philip
