Error: INACCESSIBLE BOOT RECORD

[Rod Speed]

My son has that one and he uses it for hig 250GB HD
with no problems whatsoever. It claims ATA 66/100/133.

And others have seen problems. Thats the problem with
flouting standards, can result in fangs in the arse at the
most inappropriate moments, when you discover that
what you thought was a viable backup copy isnt.

http://www.directron.com/kf23.html

I used an older model bay for that.

And thats why I suggested trying the restore
without it, to see if it was the problem.

I checked up on that unit and the vendors I talked
with claim they have had no problems with it.

And others have seen problems. One just recently in csiphs.

I have yet to talk to Directron, however, and they
have the final say as far as I am concerned because
they do an incredible volume of business. I believe
they get over 10,000 hits per day to their website.

I use the standards instead.

I will stick with PATA for now.

I will keep it as a backup machine.

Yeah, I havent had just one for many many years now.

I have a spare port on my router that I can hook
it to and do a refresh of new files now and then.

Yeah, I do that, keep the image file on one of the other PCs on the lan.

I may use that Enermax box to make archives by removing
the mirror disk each week and putting it back in a week later
to build a new mirror for archiving. I don't see any need for
dynamic mirroring - just disaster recovery backup once a week.

The reason for using that unit instead of a software approach
is that I can plug in the second drive hot and remove it hot
after the mirror is built. IOW, it is effortless - I do not have
to dedicate a special time to do the backup.

I wouldnt flout the standard myself.

I can put a

Whoops, the fault must be back |-)

 

[Rod Speed]

It is noisy. But he doesn't mind since he plays loud music on it.

And he's likely deaf by now anyway |-)

He has 3 fans - front, back and side.

The problem is that that isnt necessarily enough with the Prescotts.

You really need a pipe to the outside of the case
to ensure that the cpu gets room temp inlet air.

And even then that can still be too high in summer
if you dont have the a/c running all the time.

The CPU got to 75C when converting DVD rips for layoff and he was worried

Yeah, thats a completely obscene temp and the main reason I have
avoided the Prescotts. The Northwood I use doesnt even get to 40C at
the worse, and thats with quite high room temps that can be 34C or so.

because he literally fried two old Acer systems someone gave him to
play with. So for a few bucks he now runs cool - albeit somewhat noisy.

Yeah, I've given up on noisy systems now that the main
system is completely silent. Bit of a drug in some ways.

There is a fan speed control but Directron did not install it and he
is not going to fool with it because noise is not a problem for him.

His unit cost a bit over $50 and it does dual layer.

There isnt anything in it price wise and they are all dual layer now.

It has not caused him any problems and he has
ripped/burned several hundred DVDs with it.

Sure, but it mainly depends on the players those are played in.

The only problem he has is the conversion/compression utility he
uses most of the time can sometimes let crap in and then the copy
won't play. Usually that's caused by a library DVD. I swear the idiots
who check out library DVDs must use them to train their dog to fetch.

I won't touch SATA - too many years of being serial number one
and paying the price for it. Let the pioneers wear the arrow shirts.

I just looked it up on Directron and it's actually a bit cheaper than
the equivalent WD drive. It is ATA1333 whereas the WD is ATA100.

So I will consider switching to Samsung based on your recommendation.

That always happens when a company dominates the market.

They dont really.

I remember calling the developers at Microsoft back in the
1980s to chat about their assembler and C compiler. I found
a bug in Quick C and they confirmed it, which entitled me to
a free upgrade. I asked the guy if I could have the full C
Compiler instead and he said OK. I got a $600 compiler for free.

Those were the days when computing was really a joy.

Still is in my opinion.

 

[Bob]

Whoops, the fault must be back |-)

Perhaps. When I clone the disk for the new system I will connect it to
the IDE cable directly.

I am still mystified what happened the other day when my machine went
nuts. Perhaps if I reconstruct the events in proper order, it will be
easier to diagnose.

1) We cleaned the MB with dry compressed air about two weeks before I
had trouble.

2) The weather was getting hot & muggy - perhaps I should have run the
a/c. It was not oppressive.

3) All of a sudden the mouse started acting up. It would fail to
respond for a moment and then snap to some position where I did not
point it. Also the focused window would scroll on its own.

4) I rebooted and noticed that the POST memory check was wrong. I have
3 sticks of 128MB RAM in the machine so I always got 384MB for the
POST report. But that day I got 512MB as though one stick was not
there.

5) After I rebooted a couple of my utilities (Kerio and AdSubtract)
lost their data files. I have seen AdSubtract do that before but never
Kerio.

6) I booted the maching into DOS with Drive Image Pro because it has a
mouse driver and it was a quick and easy way to test the mouse under
DOS. The mouse did not work at all, even under DOS. That's a hardware
problem.

7) I disconnected the machine and sat it to the side while I tried to
configure one my son had cast off. The next day, after running the a/c
overnight, I cleaned the PS and MB with dry compressed air. That's
when I noticed that the PS was very dirty.

8) Because I tried to use the boot disk on another machine, I had
buggered it. That's when I ran into the problems with booting, which
is a separate problem I am not trying to figure out anymore.
Nevertheless I had to restore the archive because Kerio and AdSubtract
were corrupted.

9) After doing the IPU, I got the machine to run properly. The POST
memory check worked correctly, the mouse worked correctly and because
I had restored an earlier archive, Kerio and AdSubtract were OK.

10) Just because I am paranoid about all this, I am running a 24x24
room fan on the MB with the case off. The CPU never gets much hotter
than 40C or thereabouts, even when I am doing a defrag.

 

[Rod Speed]

Rod Speed <[email protected]> wrote

When I clone the disk for the new system
I will connect it to the IDE cable directly.

I am still mystified what happened the other day when
my machine went nuts. Perhaps if I reconstruct the
events in proper order, it will be easier to diagnose.

Yeah, I'd certainly get a handle on whether its just some quirk
of DIP and XP once you have moved to a new system, so you
can be completely confident you can restore when you need to.

The initial failure to boot I mean, not the original fault.

1) We cleaned the MB with dry compressed
air about two weeks before I had trouble.

2) The weather was getting hot & muggy - perhaps
I should have run the a/c. It was not oppressive.

3) All of a sudden the mouse started acting up. It would fail to
respond for a moment and then snap to some position where I
did not point it. Also the focused window would scroll on its own.

Thats the main thing thats hard to see how it could
be anything to do with dirt in the power supply.

4) I rebooted and noticed that the POST memory check was wrong. I
have 3 sticks of 128MB RAM in the machine so I always got 384MB for the
POST report. But that day I got 512MB as though one stick was not there.

Guess that could have been caused by the use of compressed
air on the motherboard, one dimm which wasnt properly seated
got moved a bit and was no longer seen by the bios at boot time.

5) After I rebooted a couple of my utilities (Kerio and AdSubtract) lost
their data files. I have seen AdSubtract do that before but never Kerio.

Cant say I have ever used AdSubtract.

6) I booted the maching into DOS with Drive Image Pro
because it has a mouse driver and it was a quick and easy
way to test the mouse under DOS. The mouse did not
work at all, even under DOS. That's a hardware problem.

7) I disconnected the machine and sat it to the side while I tried
to configure one my son had cast off. The next day, after running
the a/c overnight, I cleaned the PS and MB with dry compressed
air. That's when I noticed that the PS was very dirty.

8) Because I tried to use the boot disk on another machine, I had
buggered it. That's when I ran into the problems with booting,
which is a separate problem I am not trying to figure out anymore.

Yeah, floppys are a real problem. I use bootable CDs now instead.

And use the Ultimate Boot CD when testing systems too.

Nevertheless I had to restore the archive
because Kerio and AdSubtract were corrupted.

9) After doing the IPU, I got the machine to run properly.
The POST memory check worked correctly,

Maybe it either got cool enough or the extra use of compressed
air moved it back again if the connection was a tad marginal.

the mouse worked correctly and because I had restored
an earlier archive, Kerio and AdSubtract were OK.

10) Just because I am paranoid about all this, I am
running a 24x24 room fan on the MB with the case off.

Yeah, I would until the new system is up and running.

The CPU never gets much hotter than 40C or
thereabouts, even when I am doing a defrag.

Yeah, thats pretty typical for a system of that era in that config.

 

[Bob]

Yeah, I'd certainly get a handle on whether its just some quirk
of DIP and XP once you have moved to a new system

I am running Win2K and will do so with the new system for a while.
Then I will do an IPU with XP Pro.

I do not know if it's DIP causing the problem or not. One thing I did
not mention that could have direct bearing on the non-bootable issue
is the following.

I am using an old version of DIP 4 I got many years ago. I tried a new
one and it did not set up the network properly and would not do the
Didk to Disk copy from within Win2K. I fussed about Error 22 but I
could not find that particular error in the PDFs. So I abandoned it
and continue my trust old version that appears to work.

However, I also have an old version of Partition Magic 4.0 and it
fusses whenever I try to partition my hard disks. It claims that the
cylinder count is greater than 1,024 and that causes the partition to
be non-bootable. That's really lame so I got a copy of PM 8.0 but it
does the same thing. I can't believe a product like Partition Magic
can't figure out how to make a bootable partition. So I use Win2K Disk
Manager to do it.

Now here's the point. DIP uses PM's partitioner to size the target
disk when it needs to be resized. However it does not fuss about 1,024
cylinders. That could be either it does not have a problem, or is does
have a problem with creating a non-bootable target but doesn't say
anything. It does seem to copy the partition table because the target
ends up Primary and Active.

Maybe this is what has caused this non-bootable problem. The only
other time I did a restore with DIP was a long time ago and I could
have forgotten that I had to do a IPU to get the new target boot disk
to boot.

I will take a look at True Image - maybe they know what they are
doing. However this is all academic when I get my new machine with the
Enermax backup box.

 

[Rod Speed]

I am running Win2K

Yeah, just my brain fart. I dont bother to proof read my posts, too boring.

and will do so with the new system for a while.
Then I will do an IPU with XP Pro.

Yeah, XP is well worth upgrading to. I wouldnt IPU it tho,
I'd do a clean install and use the Files and Settings Transfer
Wizard in XP to migrate the settings and files to the new install.

I do not know if it's DIP causing the problem or not.

Yeah, thats what I meant by careful testing, to pin down
exactly what it was that produced a non bootable 2K.

I'd wipe the drive with something like clearhdd before doing a
restore and see if DIP does restore everything 2K needs to boot.

One thing I did not mention that could have direct
bearing on the non-bootable issue is the following.

I am using an old version of DIP 4 I got many years ago.

Thats certainly getting a tad ancient with 2K and NTFS format.

I tried a new one and it did not set up the network properly

Yeah, its not trivial to get networking working if the NIC isnt
in the list of natively supported NICs. Doable, but fiddly.

and would not do the Didk to Disk copy from within
Win2K. I fussed about Error 22 but I could not find
that particular error in the PDFs. So I abandoned it
and continue my trust old version that appears to work.

It may however not have been copying everything
back with a restore. The way to prove that is to
wipe the original drive with something like clearhdd
and see if the copy boots once you restore.

You may have managed to lose the critical bit that doesnt
get copied with the original corruption of the drive and that
may be why 2K woudnt boot after the restore.

However, I also have an old version of Partition Magic 4.0 and it
fusses whenever I try to partition my hard disks. It claims that the
cylinder count is greater than 1,024 and that causes the partition to
be non-bootable. That's really lame so I got a copy of PM 8.0 but it
does the same thing. I can't believe a product like Partition Magic
can't figure out how to make a bootable partition.

I dont normally use it, basically use True Image most
of the time now and previously used Drive Image 2002

So I use Win2K Disk Manager to do it.

I use the partitioner you get when you boot the CD when doing
a clean install if I want more than one partition on the boot drive.

Except on the test machine where I normally do the partitioning
using the XP disk manager, except when I want a larger than
32G FAT32 partition which XP refuses to create.

Now here's the point. DIP uses PM's partitioner to size the target
disk when it needs to be resized. However it does not fuss about 1,024
cylinders. That could be either it does not have a problem, or is does
have a problem with creating a non-bootable target but doesn't say
anything. It does seem to copy the partition table because the target
ends up Primary and Active.

Maybe this is what has caused this non-bootable problem.

Yeah, thats a distinct possibility.

The only other time I did a restore with DIP was a
long time ago and I could have forgotten that I had
to do a IPU to get the new target boot disk to boot.

Yeah, easy to forget the detail with stuff like that done a while ago.

I will take a look at True Image - maybe they know what they are doing.

Yeah, tho there is one major downside, since its linux
based when booting the TI CD, the support for less than
bog standard hardware can be a significant problem
and that might bite you with the enermax backup box.

However this is all academic when I get my
new machine with the Enermax backup box.

If it actually works |-)

 

[Bob]

Yeah, XP is well worth upgrading to.

My son is using it in Win2K compatibility mode.

I wouldnt IPU it tho,
I'd do a clean install and use the Files and Settings Transfer
Wizard in XP to migrate the settings and files to the new install.
OK

Yeah, thats what I meant by careful testing, to pin down
exactly what it was that produced a non bootable 2K.

I tried to clone a partition to another partition on the same drive
with True Image but it won't let me. It wants two physical disks.

I'd wipe the drive with something like clearhdd before doing a
restore and see if DIP does restore everything 2K needs to boot.

The days of DIP are numbered now that I will have the Enermax box to
do both mirroring and archiving.

Thats certainly getting a tad ancient with 2K and NTFS format.

I've had it with that crap. If True Image works, then I will use it.

Yeah, its not trivial to get networking working if the NIC isnt
in the list of natively supported NICs.

3Com 305C. The industry standard.

Doable, but fiddly.

It did not give me any reasons for failing. Typical.

You may have managed to lose the critical bit that doesnt
get copied with the original corruption of the drive and that
may be why 2K woudnt boot after the restore.

I assume the partition utility embedded in DIP is the culprit. There
are three options to make a disk clone and I chose the one which
effectively repartitions the volume.

basically use True Image most
of the time now and previously used Drive Image 2002

That's the verion of DIP I tried that screwed up the networking.

Yeah, thats a distinct possibility.

I think that's what caused the non-bottable problem.

If it actually works |-)

I said I was not going to use serial number one ever again - and here
I go again.

It must be genetic.

 

[Bob]

Any comments on this "1024 problem"?

Why would PM 8.0 not be able to deal with large drives?

Maybe my BIOS is so old (1999) that PM doesn't know how to deal with
the translation to large drives.

 

[Rod Speed]

My son is using it in Win2K compatibility mode.

Mad in my opinion, but then I have never had a
problem with the user interface stuff moving on.

I tried to clone a partition to another partition on the same drive
with True Image but it won't let me. It wants two physical disks.

Yeah, spose it does. Havent needed to do that, I only
clone partitions when migrating them between drives.

You could clone a partition to another on the same
physical drive by creating an image file and then
restoring that to a different partition on the drive.

The days of DIP are numbered now that I will have
the Enermax box to do both mirroring and archiving.

Well, claims it can, anyway |-)

I've had it with that crap. If True Image works, then I will use it.

Yeah, that was my reaction too. I attempted to use Ghost 9
when I changed the boot drive on the main system to a bigger
drive and gave up in disgust. TI does a much better job of that.

3Com 305C. The industry standard.

Its got a few versions tho with some real quirks with some versions.

Not insurmountable quirks, just a damned nuisance.

It did not give me any reasons for failing. Typical.

Yeah, thats the reason its fiddly, you dont get much
useful in the way of error messages, it just doesnt work.

Ghost 2003 is even worse with NICs that arent natively supported.
The wizard is damned cryptic and the help close to useless.

I assume the partition utility embedded in DIP is the culprit.
There are three options to make a disk clone and I chose
the one which effectively repartitions the volume.

That's the verion of DIP I tried that screwed up the networking.

OK, I do recall I had some hassles with it with a 3Com 305C NIC.
Basically the default NIC config was the problem. Easy enough to
fix using the 3com ute to reconfigure the NIC.

I think that's what caused the non-bottable problem.

Yeah, thats certainly the most likely possibility.

I said I was not going to use serial number
one ever again - and here I go again.

It must be genetic.

A Jap would have the decency to disembowel itself |-)

 

[Rod Speed]

Any comments on this "1024 problem"?

Nope. I dont use it much.

Why would PM 8.0 not be able to deal with large drives?

It handles large drives fine. You shouldnt have a cylinder
count bigger than 1024 if you use an AUTO drive type.

Maybe my BIOS is so old (1999) that PM doesn't
know how to deal with the translation to large drives.

Its more likely that someone setup the drive with
a cylinder value bigger than 1024 and thats still visible
in the MBR and thats what PM is complaining about.

There's geometry details in the MBR.

 

[Bob]

Mad in my opinion, but then I have never had a
problem with the user interface stuff moving on.

The native XP GUI is too confusing for me. It took a while to catch on
to Win2K GUI so I do not want to lose that investment. MS just loves
to hide things with obscure names so once I learn where they are I do
not want to have to learn them all over again.

 

[Bob]

You shouldnt have a cylinder
count bigger than 1024 if you use an AUTO drive type.

I am using AUTO.

Its more likely that someone setup the drive with
a cylinder value bigger than 1024 and thats still visible
in the MBR and thats what PM is complaining about.

Win2K doesn't complain.

 

[Jeremy Boden]

Any comments on this "1024 problem"?

Why would PM 8.0 not be able to deal with large drives?

Maybe my BIOS is so old (1999) that PM doesn't know how to deal with
the translation to large drives.

Possibly.
I assume the disk is of similar vintage?

This will be the more than XXXX cylinders problem, where XXXX could be
various numbers (max 1024). The problem is that older BIOS's cannot
access areas of disk expressed in CHS format. Hence your boot process
would fail if your boot loader program is beyond 1024 "Cylinders".

PM should be able to copy non-bootable partitions of any size. In any
case it makes sense to reserve a partition for your OS - which would
probably get round any cylinder problems?

 

[Bob]

Possibly.
I assume the disk is of similar vintage?

Yes, maybe a couple years older.

For my new system I am getting all new parts.

This will be the more than XXXX cylinders problem, where XXXX could be
various numbers (max 1024).

On mine, the number is 1024.

The problem is that older BIOS's cannot
access areas of disk expressed in CHS format. Hence your boot process
would fail if your boot loader program is beyond 1024 "Cylinders".

The BIOS is an Award of August 1999 vintage.

PM should be able to copy non-bootable partitions of any size.

It can copy bootable partitions of any size but when it does it makes
them non-bootable.

In any case it makes sense to reserve a partition for your OS - which would
probably get round any cylinder problems?

I assume by that you mean to make the boot partition small enough to
get around the limitations referenced above. That's what Dell used to
do IIRC. I do not care to do that because I want everything on one
partition for ease in making backups.

I assume that I won't have this problem with my new machine because
the BIOS and hard drive will be up to date.

Actually the non-bootable problem could be a blessing in disguise
because it required me to do the IPU which "rebuilt" my Win2K system,
for what that's worth.

 

[Rod Speed]

The native XP GUI is too confusing for me.

It certainly does take a bit of getting used to, like any change
does, but on the whole the change is to a better approach.

It took a while to catch on to Win2K GUI
so I do not want to lose that investment.

You'll eventually find that the 2K GUI wont be supported anymore.

MS just loves to hide things with obscure
names so once I learn where they are I do
not want to have to learn them all over again.

Things always change over time. In spades with computing.

Nothing MS specific there.

 

[Rod Speed]

I am using AUTO.

You sure you were when that drive was first installed tho ?

Win2K doesn't complain.

Yeah, PM is more cautious because when crunching
partitions the consequences of bad data can be very
catastrophic indeed, like nothing visible anymore.

 

[Bob]

Things always change over time. In spades with computing.

UNIX hasn't changed since the days of Bell Labs.

 

[Bob Eager]

UNIX hasn't changed since the days of Bell Labs.

Shows how little you know. I was using it 29 years ago, and I still use
it. It's radically different.

Even that jumped-up UNIX wannabe [1] isn't too bad.

[1] Calls itself Linux.

 

[Bob]

You sure you were when that drive was first installed tho ?

It was set up that way when it was built and no one has ever changed
it.

Yeah, PM is more cautious because when crunching
partitions the consequences of bad data can be very
catastrophic indeed, like nothing visible anymore.

Maybe Win2K works and PM doesn't.

Here's the partition info dump. The drive is a WD 30GB divided into
nearly equal partitions. The second partition contains a DIP disk copy
of the first.

Disk Geometry Information for Disk 1: 3649 Cylinders, 255 Heads,
63 Sectors/Track
System PartSect # Boot BCyl Head Sect FS ECyl Head
Sect StartSect NumSects

0 0 80 0 1 1 07 1023 254
63 63 29,302,497
Info: End C,H,S values were large drive placeholders.
Actual values are:
0 0 80 0 1 1 07 1823 254 63 63
29302497
0 1 00 1023 0 1 07 1023 254
63 29,302,560 29,318,625
Info: Begin C,H,S values were large drive placeholders.
Info: End C,H,S values were large drive placeholders.
Actual values are:
0 1 00 1824 0 1 07 3648 254 63 29302560
29318625

Partition Information for Disk 1: 28,623.6 Megabytes
Volume PartType Status Size MB PartSect #
StartSect TotalSects

C: NTFS Pri,Boot 14,307.9 0 0
63 29,302,497
F: NTFS Pri 14,315.7 0 1
29,302,560 29,318,625

Boot Record for drive C: (Drive: 1, Starting sector: 63, Type: NTFS)

1. Jump: EB 52 90
2. OEM Name: NTFS
3. Bytes per Sector: 512
4. Sectors per Cluster: 1
5. Reserved Sectors: 0
6. Number of FATs: 0
7. Root Dir Entries: 0
8. Total Sectors: 0
9. Media Descriptor: 0xF8
10. Sectors per FAT: 0
11. Sectors per Track: 63 (0x3F)
12. Number of Heads: 255 (0xFF)
13. Hidden Sectors: 63 (0x3F)
14. Total Sectors (>32MB): 0 (0x0)
15. Unused: 0x80008000
16. Total NTFS Sectors: 29302496
17. MFT Start Cluster: 14870173
18. MFT Mirror Start Clust: 4192933
19. Clusters per FRS: 2
20. Clusters per Index Blk: 8
21. Serial Number: 0x00000000427D8D6B
22. Checksum: 0 (0x0)
23. Boot Signature: 0xAA55

Boot Record for drive F: (Drive: 1, Starting sector: 29,302,560,
Type: NTFS)

1. Jump: EB 52 90
2. OEM Name: NTFS
3. Bytes per Sector: 512
4. Sectors per Cluster: 1
5. Reserved Sectors: 0
6. Number of FATs: 0
7. Root Dir Entries: 0
8. Total Sectors: 0
9. Media Descriptor: 0xF8
10. Sectors per FAT: 0
11. Sectors per Track: 63 (0x3F)
12. Number of Heads: 255 (0xFF)
13. Hidden Sectors: 29302560 (0x1BF1F20)
14. Total Sectors (>32MB): 0 (0x0)
15. Unused: 0x80008000
16. Total NTFS Sectors: 29318624
17. MFT Start Cluster: 14870173
18. MFT Mirror Start Clust: 4192933
19. Clusters per FRS: 2
20. Clusters per Index Blk: 8
21. Serial Number: 0x000000004284B22F
22. Checksum: 0 (0x0)
23. Boot Signature: 0xAA55

 

[Rod Speed]

UNIX hasn't changed since the days of Bell Labs.

Bullshit, most modern versions have GUIs.

And even the shells have changed quite a bit since the days of the bell labs.

Knoppix for example is nothing like the days of Bell Labs.