ME Reinstallation (Clean Install) Problems

Good point - FDISK /STATUS should show whether the disk is formatted
correctly or not.

--
Noel Paton (MS-MVP 2002-2004, Win9x)

Nil Carborundum Illegitemi
http://www.btinternet.com/~winnoel/millsrpch.htm

Please read http://dts-l.org/goodpost.htm on how to post messages to NG's
or
http://www.microsoft.com/presspass/f.../Mar27pmvp.asp

"BarryG" wrote in message
...
snip
I know that the drive is good as I've used it as a
slave in point 1 as well as on another XP machine.
snip

I find it hard to believe that it's the WD software, as I
used the same
software to format a DD for XP, and it worked fine.


Perchance, if you are using the hard disk in a XP machine,
is it formatted as NTFS? If so, then WinME can't see it
when booted from a WinME startup disk. Maybe ramdisk is
then C: drive, and CD-ROM becomes D: drive?
You really need to Fdisk the hard disk with WinME's fdisk,
and format with with WinME's format to really be sure you
get fat32, which WinME can read.

Good luck,
BarryG




In article tKjBc.596
,
says...

I had a processor die a few weeks ago, so I decided to
install a new
processor
and motherboard. Of course this played havoc with the
operating system and
made it very unstable. It'us useable, but not stable at
all.


.

Noel and Barry, you just may be on to something here. I have been formatting the drive on an XP
machine, but I have specifically been formatting as an ME DD (or at least I think I have. I've
been farting around with this long enough that there's not really much that I am sure of!)

1) Ramdisk is C: and the CD is D: -- as you suggested (when using an ME start floppy in the
non-working machine.

2)In the non-working machine set up as a single drive: no recognition of the boot capability --
using an ME start flopp and ME fdisk does not recognize the drive . When I flip it into slave
mode and use another ME DD to boot the machine, only the first partition of the trouble DD is
recognized

3)In the working XP machine: full recognition of the partitions (as set up as a slave).

4)For the heck of it one time I did a bootable c: in XP format and did the equivalent of a copy
*.* from the XP c: to the DD I've been having trouble with. So the thing should have booted as an
XP machine. Took it to the machine I've been having trouble with, set it up as a single drive ,
-- and the da*n thing still did not recognize the c: boot sectors.

4a)After doing 4 above, I took the thing back to the XP machine and put the drive in as a single
drive -- and the machine DID NOT recognize a bootable c: (I did not check this on the XP machine
before doing 4 above)

5)BIOS is set up with SMART HD enabled and I've run all tests on the drive, and everything is ok.

6)Ran a current AV on the drive and although there are no viruses, I did see something strange --
2 master boot records and 8 boot records on a 3 partition drive.

7)One other strange thing: reference 4 above -- the non-working machine is adding one file to the
drive WIN386.SWP at 47,200 bytes.

I've just formatted the DD again -- this time as non bootable again in ME format. I'll try ME
fdisk again.

Any other suggestions? Thanks for your help



Perchance, if you are using the hard disk in a XP machine,
is it formatted as NTFS? If so, then WinME can't see it
when booted from a WinME startup disk. Maybe ramdisk is
then C: drive, and CD-ROM becomes D: drive?
You really need to Fdisk the hard disk with WinME's fdisk,
and format with with WinME's format to really be sure you
get fat32, which WinME can read.

Good luck,
BarryG




In article tKjBc.596
,
says...

I had a processor die a few weeks ago, so I decided to
install a new
processor
and motherboard. Of course this played havoc with the
operating system and
made it very unstable. It'us useable, but not stable at
all.


.

Well, after all the dailogue, I guess I'm convinced that the problem involves my switching back
and forth bewteen the XP and ME formats - and the incompatability of XP's DD formatting vs that
of ME.

Problem is that I can't use ME fdisk to fix the issue as it does not recognize the DD. I
apparently can't use Western Digital's current software as it does not seem to have worked so
far.

So I, went back to three year old WD software that I've used to set up ME DDs in the past. I'm
running a BIOS check on the DD now. The old WD software will load EX-BIOS on the DD and partition
the thing. The only continuing issue is that this fix is SLOOOOOWWWWWW. It's taking the old
software about 2 seconds a cylinder -- so, if I'm lucky it might be done tomorrow!

Thanks for putting up with my insanity!

Oh, I forgot to make an observation: It sure would have been easier for me to go out and get a
new 120 gig drive for 50 bucks as I've sure put a lot more than 50 bucks of labor into this
effort.

On Thu, 24 Jun 2004 17:04:35 +0100, "Mike M"

1) Good. This shows that the system is OK and the problem solely related to
the new hard disk.

Oversnip and missed earlier posts may account for this, but from what
I've read of this thread, I haven't seen enough to exclude bad
hardware other than HD (i.e. RAM, overclocking, heating issues, or
problems within "the untestables").

2) Why use WD's LifeGuard rather than fdisk and format using a Win Me boot
floppy?

If HD 99G, FDisk - even WinME or "fixed" one for older Win9x - will
not be able to properly display capacities over 99G or (more
significantly) allow such capacites to be input.

For that, I use BING from www.bootitng.com

5) I don't see what the attached disk drive has to do with the bios memory
count which only normally takes place when the PC is first switched on and
before any attempt has been made to access the hard disk.

Yep. Makes me wonder about a crash that ate CMOS settings, defaulting
back to "fast/quiet boot". Manual setup that changed that BIOS/CMOS
default may also have done other changes to keep the system stable
with overclocking or sub-spec RAM, and if those settings were lost at
the same time, then you may end up with porridge.

I'd do http://cquirke.mvps.org/9x/bthink.htm before trying to
(re-)install the OS again. Clearly doing so for the nth time is
unlikely to be time well spent :-(




-------------------- ----- ---- --- -- - - - -
No, perfection is not an entrance requirement.
We'll settle for integrity and humility
-------------------- ----- ---- --- -- - - - -

cquerke, thanks for your post. I'm still nowhere with this. I've installed the WD software on a
drive with a working ME OS. Loaded ME then went into the software and VERY SPECIFICALLY formatted
the problem DD as bootable and with a FAT32 designation. WD's software then let's you do a
complete copy of the working c drive to what will be the first partition of the new drive. When
it was done, I made the problem drive the c drive, turned the machine on -- and the machine did
not view the DD as bootable. This is the same problem I've had. I ran fdisk from an ME start
floppy, and again fdisk could not access the problem DD. The thing is almost acting like it
refuses to be reformatted - but no error messages show up.



Oversnip and missed earlier posts may account for this, but from what
I've read of this thread, I haven't seen enough to exclude bad
hardware other than HD (i.e. RAM, overclocking, heating issues, or
problems within "the untestables").



I'm using half a gig of kingston 2700 ddr, have not overclocked the cpu, have an extra capacity
heart sink and fan on the cpu. Have set an emergency overheat shutdown in BIOS -- and have not
had a shutdown. As I said in my first post a very long time ago, we're dealing with a new
motherboard and an AMD 2500+ cpu. Plus a different DD loaded with the ME OS boots fine.



If HD 99G, FDisk - even WinME or "fixed" one for older Win9x - will
not be able to properly display capacities over 99G or (more
significantly) allow such capacites to be input.




I'm dealing with a 30 gig Western Digital DD that's 3 years old, but has passed every SMART test
there is.




Yep. Makes me wonder about a crash that ate CMOS settings, defaulting
back to "fast/quiet boot". Manual setup that changed that BIOS/CMOS
default may also have done other changes to keep the system stable
with overclocking or sub-spec RAM, and if those settings were lost at
the same time, then you may end up with porridge.



I did have to reset the memory jumpers on the motherboard because the machine was misrecognizing
the CPU and memory speed as an AMD 1900+ and memory speed of 133.; In reality it was AMD 2500+
and memory speed of 166.

I'd wonder about other hardware, CMOS settings etc, too -- except for one thing: If I take
another old DD (and old 6 gig WD that was built in '97) that has a working ME OS on it, the
system does boot. ME does start. Everything works with this other DD.

That leads me ot the only conclusion that I can make: this is a DD problem. I just don't
understand what it is.

I'm out of here for the weekend. I'll take a look at any responses on Monday. Again thanks to
those of you that have ventured opinions.

When you made the new disk the master, did you remember to reset the
jumpers??

--
Noel Paton (MS-MVP 2002-2004, Win9x)

Nil Carborundum Illegitemi
http://www.btinternet.com/~winnoel/millsrpch.htm

Please read http://dts-l.org/goodpost.htm on how to post messages to NG's
or
http://www.microsoft.com/presspass/f.../Mar27pmvp.asp

"RN" wrote in message
.. .
cquerke, thanks for your post. I'm still nowhere with this. I've installed
the WD software on a
drive with a working ME OS. Loaded ME then went into the software and VERY
SPECIFICALLY formatted
the problem DD as bootable and with a FAT32 designation. WD's software
then let's you do a
complete copy of the working c drive to what will be the first partition
of the new drive. When
it was done, I made the problem drive the c drive, turned the machine
on -- and the machine did
not view the DD as bootable. This is the same problem I've had. I ran
fdisk from an ME start
floppy, and again fdisk could not access the problem DD. The thing is
almost acting like it
refuses to be reformatted - but no error messages show up.



Oversnip and missed earlier posts may account for this, but from what
I've read of this thread, I haven't seen enough to exclude bad
hardware other than HD (i.e. RAM, overclocking, heating issues, or
problems within "the untestables").



I'm using half a gig of kingston 2700 ddr, have not overclocked the cpu,
have an extra capacity
heart sink and fan on the cpu. Have set an emergency overheat shutdown in
BIOS -- and have not
had a shutdown. As I said in my first post a very long time ago, we're
dealing with a new
motherboard and an AMD 2500+ cpu. Plus a different DD loaded with the ME
OS boots fine.



If HD 99G, FDisk - even WinME or "fixed" one for older Win9x - will
not be able to properly display capacities over 99G or (more
significantly) allow such capacites to be input.




I'm dealing with a 30 gig Western Digital DD that's 3 years old, but has
passed every SMART test
there is.




Yep. Makes me wonder about a crash that ate CMOS settings, defaulting
back to "fast/quiet boot". Manual setup that changed that BIOS/CMOS
default may also have done other changes to keep the system stable
with overclocking or sub-spec RAM, and if those settings were lost at
the same time, then you may end up with porridge.



I did have to reset the memory jumpers on the motherboard because the
machine was misrecognizing
the CPU and memory speed as an AMD 1900+ and memory speed of 133.; In
reality it was AMD 2500+
and memory speed of 166.

I'd wonder about other hardware, CMOS settings etc, too -- except for one
thing: If I take
another old DD (and old 6 gig WD that was built in '97) that has a working
ME OS on it, the
system does boot. ME does start. Everything works with this other DD.

That leads me ot the only conclusion that I can make: this is a DD
problem. I just don't
understand what it is.

I'm out of here for the weekend. I'll take a look at any responses on
Monday. Again thanks to
those of you that have ventured opinions.

In article ,
says...

When you made the new disk the master, did you remember to reset the
jumpers??


Thanks Noel. Yes, I set the jumpers correclty. When I use as a single DD, I
always take off all jumpers. For WD DDs a master is set on the middle set of
pins and a slave is set on the set of pins right of center,

For whatever reason I believe that the disk is not reformatting completely.
The thing was set up three years ago as a 4 partition non bootable ME DD and
was used as a slave.

Now I'll flash forward to a few days ago, after multiple attempts to reformat
the thing, I was at what I thought was 3 partitions. I ran a Norton AV on it,
found no virus activity, but did find 2 master boot records and 8 boot
records. I can understand 1 and 3, but 2 and 8 on a freshly partitioned 3
partition disk? That does not make sense.

There's often a backup set made - but 8 BR's on a 3-partition disk does
sound strange.

I also would no;t trust anything that a Norton application told me about
anything to do with windows - they have enough problems identifying problems
with their own files , and are totally unreliable as regards anything else.


--
Noel Paton (MS-MVP 2002-2004, Win9x)

Nil Carborundum Illegitemi
http://www.btinternet.com/~winnoel/millsrpch.htm

Please read http://dts-l.org/goodpost.htm on how to post messages to NG's
or
http://www.microsoft.com/presspass/f.../Mar27pmvp.asp

"RN" wrote in message
...
In article ,
says...

When you made the new disk the master, did you remember to reset the
jumpers??


Thanks Noel. Yes, I set the jumpers correclty. When I use as a single DD,
I
always take off all jumpers. For WD DDs a master is set on the middle set
of
pins and a slave is set on the set of pins right of center,

For whatever reason I believe that the disk is not reformatting
completely.
The thing was set up three years ago as a 4 partition non bootable ME DD
and
was used as a slave.

Now I'll flash forward to a few days ago, after multiple attempts to
reformat
the thing, I was at what I thought was 3 partitions. I ran a Norton AV on
it,
found no virus activity, but did find 2 master boot records and 8 boot
records. I can understand 1 and 3, but 2 and 8 on a freshly partitioned 3
partition disk? That does not make sense.

On Sat, 26 Jun 2004 00:21:43 GMT, (RN)

cquerke, thanks for your post. I'm still nowhere with this. I've installed the WD software on a
drive with a working ME OS.

I'm nervous about using vendorware for partitioning and formatting,
unless I am COAB certain that it doesn't insert non-standard code into
the pre-OS boot axis (i.e. the MBR).

If it does, then the way a HD boot and a non-HD boot will see the HD
may vary, and that will chew your butt hugely if you ever need to do
data recovery, formal virus check, or other hands-off maintenace. If
anything - such as an OS install, or av clean-up of MBR infector, or
side-effect of some other maintenace - replaces the "special" MBR with
what is after all supposed to be standard system code, then Snap!

Loaded ME then went into the software and VERY SPECIFICALLY formatted
the problem DD

Diskette Drive or Hard Drive?

as bootable and with a FAT32 designation. WD's software then let's you do a
complete copy of the working c drive to what will be the first partition of the new drive. When
it was done, I made the problem drive the c drive, turned the machine on -- and the machine did
not view the DD as bootable. This is the same problem I've had.

There are two points at which pre-file-system code has to be present,
plus one additional setting. These a

1) MBR (Master Boot Record); system-level boot code
2) Active byte set within partition table for that partition
3) PBR (Partition Boot Record); OS-level boot code

If any of those are missing, then no boot.

FDisk won't automatically do (2) unless the primary is created
"dummy-style" as all of the HD, and it will refuse to set the byte at
all unless it's the HD in the PC and thus "bootable".

I'd not expect the WD sware to inherit (or mimic) this limitation, but
it's possible, so check that out.

How to fix (1):
- new HD as first HD on system
- use FDisk /MBR to copy standard system boot code to the new HD

How to fix (2):
- new HD as first HD on system
- FDisk "2" to set primary partition active

How to fix (3):
- new HD as first HD on system
- correct OS version diskette boot with the Sys program on it
- use Sys C: to copy in PBR, IO.SYS etc. to "make bootable"

I ran fdisk from an ME start floppy, and again fdisk could not access the
problem DD. The thing is almost acting like it refuses to be reformatted

Ah, hello! Look out for settings - CMOS most likely, but also
concievably jumpers - that "protect" the HD, make it read-only, or
enable "antivirus protection" that walls out writes to MBR.

Also, remember that FDisk may only write changes to HD when you exit
FDisk (so don't reboot without exiting FDisk) and that Format doesn't
re-read the partition structure from HD before formatting (so always
reboot after FDisking or partitioning, before Formatting)

Some HD defect patterns (or interaction hassles) can cause this sort
of "I write changes but don't see them" problems. Incorrect geometry
can cause changes to be written to the wrong place, so that they
appear not to have stuck as well.

Make sure CMOS is set to use HD as LBA, not Auto, Large, CHS or
Normal, else you may have geometry/addressing issues too.

Oversnip and missed earlier posts may account for this, but from what
I've read of this thread, I haven't seen enough to exclude bad
hardware other than HD (i.e. RAM, overclocking, heating issues, or
problems within "the untestables").

I'm using half a gig of kingston 2700 ddr, have not overclocked the cpu, have an extra capacity
heart sink and fan on the cpu. Have set an emergency overheat shutdown in BIOS -- and have not
had a shutdown. As I said in my first post a very long time ago, we're dealing with a new
motherboard and an AMD 2500+ cpu.

I still don't see RAM diags there, though this isn't the sort of
pattern I'd expect with that kind of problem. You did say WD's diags
are happy with the HD. May be worthwhile checking the FAQ section of
your mobo's web site, in case there's an "issue" there; I don't like
suggesting BIOS updates, but it's a possible requirement.

New hardware certainly does fail, and sometimes in the first few weeks
(if manufacturing defects bite).

I'm dealing with a 30 gig Western Digital DD that's 3 years old, but has
passed every SMART test there is.

Ah, it's a 30G. Well, that should be well within any sort of capacity
limits for any new motherboard - but it may be that the WD installware
predates WinME and doesn't work properly with it?

Thinking: 30G is close to 32G limit, and if part of a range that goes
over that limit, may ship with installware that uses "special" MBR
code to override BIOS in case BIOS isn't 32G OK.

Unlikely, given that when the 32G limit bites, it usual;ly does so on
IDE detection - so the driver never gets to run. More likely that
approach would be used to overcome an older 8G limitation.

Yep. Makes me wonder about a crash that ate CMOS settings, defaulting
back to "fast/quiet boot". Manual setup that changed that BIOS/CMOS
default may also have done other changes to keep the system stable
with overclocking or sub-spec RAM, and if those settings were lost at
the same time, then you may end up with porridge.

I did have to reset the memory jumpers on the motherboard because the machine was misrecognizing
the CPU and memory speed as an AMD 1900+ and memory speed of 133.; In reality it was AMD 2500+
and memory speed of 166.

RAM speed of 166MHz sounds rather non-standard. What's the PCI speed?
If it's over 33MHz, you may well see problems there.

Did you have to change jumpers for CPU voltage? Seems odd, to still
have to use jumpers to set up a CPU in 2004.

I'd wonder about other hardware, CMOS settings etc, too -- except for one thing: If I take
another old DD (and old 6 gig WD that was built in '97) that has a working ME OS on it, the
system does boot. ME does start. Everything works with this other DD.

That's still a shrug - when you break the surface of the digital level
of abstraction (as flakyness at the analog level does) then very
strange things can happen. For example, a cap that can ramp up a wave
pulse in time for the clock with one HD in place may fail to do so if
the other HD pulls a bit more juice.

That leads me ot the only conclusion that I can make: this is a DD problem. I just don't
understand what it is.

One quick guess would be CMOS address type not set to LBA. Another
quick guess would be "special" MBR code imposed by WD's sware.


--------------- ----- ---- --- -- - - -
Never turn your back on an installer program
--------------- ----- ---- --- -- - - -