Question about backups with Ghost

[Jeff]

Question for the experts.....

I have never used Ghost before so here is my setup.

Win xp sp2 with two internal sata drives each in a removeable drawer.

System drive is C

Storage drive is D

What I'd love to be able to do is the following:

Back up Drive C to Drive D so that if something gets screwed up i can simply
remove Drive C and then the "D" drive would become an exact replacement
which would pick up where "C" left off.

Is this possible?

Can I do Incremental backups nightly which would only backup files and
folders which have changed without copying the entire contents?

If thiis is indeed possible....how can a backup drive which is a clone of
the original boot at the same time as the system drive, how does windows
know the difference?

TIA

jeff

 

[Peter]

No. It is not possible.
First of all you schedule your "backup" nightly, so when you decide to swap
them, it is not the same time as you "left off your C drive".
Secondly, your D drive must be invisible to OS when C drive is active, yet
be able to start normally when swapped. The only way to achieve that is to
have it powered down except backup. Unless you put a programmable switch on
D drive power line, I do not see how can one achieve that.

 

[Bob Davis]

Question for the experts.....

I have never used Ghost before so here is my setup.

Win xp sp2 with two internal sata drives each in a removeable drawer.

System drive is C

Storage drive is D

What I'd love to be able to do is the following:

Back up Drive C to Drive D so that if something gets screwed up i can
simply remove Drive C and then the "D" drive would become an exact
replacement which would pick up where "C" left off.

Is this possible?

Can I do Incremental backups nightly which would only backup files and
folders which have changed without copying the entire contents?

If thiis is indeed possible....how can a backup drive which is a clone of
the original boot at the same time as the system drive, how does windows
know the difference?

I used to do this on my old Win98SE machine, and never had a problem. I
would keep D: in the machine at all times, then clone C: to D: once per
week, doing incremental backups of critical files via batch files in the
interim. These batch files were run once per hour by Task Scheduler.

But XP is a different animal, and I read caveats that no second drive
containing the OS should be in the same system while running any NT-based
OS. When I upgraded to XP in 2003, I decided to discontinue this procedure
after talking with M$ tech support. They suggested it might not be a good
idea.

Now I have five old but working IDE HD's in mobile racks that I rotate for
weekly cloning. If I need a file from one of those HD's, I insert it in my
USB2 mobile rack and extract it. This introduces this clone to the system,
but I've never had a problem. There have also been times when I accidently
left it in and running for hours at a time, also with no ill effects.
I use Ghost 2003, the DOS version that boots from a floppy or CD outside
Windows.

I'm still running the batch files once per hour, but D: now contains copies
of critical folders, like business databases. So if I have a failure of C:,
I can boot from my most recent clone and be up and running, missing only the
files created or changed since that clone was made. Copying the files
dutifully backed up to D: in the interim back to C: restores the status quo.
Program updates and new installs are not covered by this procedure, but
that's something I'll deal with in due time.

 

[Jeff]

No. It is not possible.
First of all you schedule your "backup" nightly, so when you decide to
swap
them, it is not the same time as you "left off your C drive".
Secondly, your D drive must be invisible to OS when C drive is active, yet
be able to start normally when swapped. The only way to achieve that is to
have it powered down except backup. Unless you put a programmable switch
on
D drive power line, I do not see how can one achieve that.

If the second drive was an "esata" external sata (treated as internal to the
OS) would this work since it could be disconnected?

Also, once the failing "c" drive was disconnected, would the esata boot
normally?

Do incremental backups with Ghost (dont quite know which version to use yet)
replace folders which have changed no matter if they are larger or smaller?

What i mean to say, is, if a folder is less than size because some files in
that folder have been deleted, will the cloned copy in an incremental back
up show the reduced size of that folder?

TIA

jeff

 

[Rod Speed]

I have never used Ghost before so here is my setup.

I dont think Ghost is the best product for
this sort of use, True Image is much better.

Win xp sp2 with two internal sata drives each in a removeable drawer.

System drive is C

Storage drive is D

What I'd love to be able to do is the following:

Back up Drive C to Drive D so that if something gets screwed up i can
simply remove Drive C and then the "D" drive would become an exact
replacement which would pick up where "C" left off.

Is this possible?
Yes.

Can I do Incremental backups nightly which would only backup files and
folders which have changed without copying the entire contents?

Nope. xxclone can do that tho.

If thiis is indeed possible....how can a backup drive which is a clone of
the original boot at the same time as the system drive, how does windows
know the difference?

Thats controlled by which drive you specify
to boot off in the motherboard bios.

Its also possible to setup a more complicated config
using the standard xp boot manager which allows you
to specify which drive you want to boot off at boot
time. Little bit fiddly to setup, but quite doable.

The fundamental question is tho whether you really
need to be able to boot off the spare so quickly.

 

[Rod Speed]

No. It is not possible.

Yes it is.

First of all you schedule your "backup" nightly, so when you decide
to swap them, it is not the same time as you "left off your C drive".

Want to try that again in english ?

Secondly, your D drive must be invisible to OS when C drive is active,

No it doesnt.

yet be able to start normally when swapped. The only way
to achieve that is to have it powered down except backup.

Wrong. All you actually need to do is never boot the D
drive with the C drive visible. You only need to do that
when the C drive has failed, so that is no big deal at all.

Unless you put a programmable switch on D drive
power line, I do not see how can one achieve that.

Trivial, just have them both powered all the time, dont
boot off the D drive with the C drive visible. Just power
down the C drive manually before booting the D drive.
No big deal since by definition you have to be there
to specify booting off the D drive anyway.

 

[Rod Speed]

I used to do this on my old Win98SE machine, and never had a problem.
I would keep D: in the machine at all times, then clone C: to D: once
per week, doing incremental backups of critical files via batch files in
the interim. These batch files were run once per hour by Task Scheduler.

But XP is a different animal, and I read caveats that no second drive
containing the OS should be in the same system while running any NT-based
OS.

That is just plain wrong. All you have to do is
avoid BOOTING the D drive with the C drive visible.

When I upgraded to XP in 2003, I decided to discontinue
this procedure after talking with M$ tech support. They suggested it
might not be a good idea.

They're wrong.

Now I have five old but working IDE HD's in mobile racks that I
rotate for weekly cloning. If I need a file from one of those HD's,
I insert it in my USB2 mobile rack and extract it. This introduces
this clone to the system, but I've never had a problem.

You wont as long as you dont boot it.

There have also been times when I accidently left it in and running for
hours at a time, also with no ill effects.

You wont get any as long as you dont boot it.

I use Ghost 2003, the DOS version that boots from a floppy or CD outside
Windows.

True Image is much better and you dont have
to boot from floppy or CD if you dont want to.

I'm still running the batch files once per hour, but D: now contains
copies of critical folders, like business databases. So if I have a
failure of C:, I can boot from my most recent clone and be up and
running, missing only the files created or changed since that clone was
made. Copying the files dutifully backed up to D: in the interim back to
C: restores the status quo. Program updates and new installs are not
covered by this procedure, but that's something I'll deal
with in due time.

His approach gets it booted off the D drive much more
quickly. The only thing he has to do is unplug or depower
the C drive for the first boot of the D drive. He can plug
it back in again or repower it after XP has booted and
rebooted once and can get any files that have changed
since the last backup off the C drive in that config too
if the C drive isnt too bad and still spins up and mounts.

 

[Andy]

Question for the experts.....

I have never used Ghost before so here is my setup.

Win xp sp2 with two internal sata drives each in a removeable drawer.

System drive is C

Storage drive is D

What I'd love to be able to do is the following:

Back up Drive C to Drive D so that if something gets screwed up i can simply
remove Drive C and then the "D" drive would become an exact replacement
which would pick up where "C" left off.

Is this possible?

Can I do Incremental backups nightly which would only backup files and
folders which have changed without copying the entire contents?

If thiis is indeed possible....how can a backup drive which is a clone of
the original boot at the same time as the system drive, how does windows
know the difference?

When Windows sees two identical drives, it changes the disk signature
of the second one. So if you want to be able to boot the clone, save a
copy of the orignal drive's MBR. When it becomes time to boot the
clone, replace the clone's MBR with the one you saved. Then you'll be
able to boot the clone.

 

[Rod Speed]

When Windows sees two identical drives, it changes the disk signature
of the second one. So if you want to be able to boot the clone, save a
copy of the orignal drive's MBR. When it becomes time to boot the
clone, replace the clone's MBR with the one you saved. Then you'll be
able to boot the clone.

You dont need to do that to boot the clone.

 

[GTS]

Question for the experts.....

I have never used Ghost before so here is my setup.

Win xp sp2 with two internal sata drives each in a removeable drawer.

System drive is C

Storage drive is D

What I'd love to be able to do is the following:

Back up Drive C to Drive D so that if something gets screwed up i can simply
remove Drive C and then the "D" drive would become an exact replacement
which would pick up where "C" left off.

Is this possible?

Can I do Incremental backups nightly which would only backup files and
folders which have changed without copying the entire contents?

If thiis is indeed possible....how can a backup drive which is a clone of
the original boot at the same time as the system drive, how does windows
know the difference?

TIA

jeff

You could go about this a different way - I use Ghost weekly to make an
image of the C: drive and save it (compressed) to the D: drive. As I only
use the C: drive for XP and some core apps the resulting image is only 4.5gb
and fits easily onto DVD for extra security.
Incidentally I use Ghost 2003 DOS executable which is only 900 Kb, but does
not allow incremental backups - but a whole drive can be imaged in a few
minutes.
You could also make an image of the D: drive and store it on C:
Advantage of images is that you get to actually use the drives for other
things rather then just being copies of another drive!
Ghost images work fine - if you ever have a drive fail you just pop in a new
one and create drive form image, and in a few minutes you get a clone of
your old drive back.
I only use disk-to-disk cloning when upgrading say to a bigger or faster
drive - more of a one-off clone so both drives are not sitting in the PC at
the same time - though I have done this without any issues.
Another method for cloning drives is to use a removeable drive caddy -
£15ish - and just stick in the backup drive when you want to do a backup -
advantage then is that drive can be kept in another room / building so even
if a real disaster strikes you have a copy.
GTS

 

[Timothy Daniels]

When Windows sees two identical drives, it changes the disk signature
of the second one. So if you want to be able to boot the clone, save a
copy of the orignal drive's MBR. When it becomes time to boot the
clone, replace the clone's MBR with the one you saved. Then you'll be
able to boot the clone.

Any utility that will clone a drive will also copy the MBR - either
automatically or as a user-indicated option. Wanting a clone
that doesn't have an MBR that will boot it is an exceptional - even
pathological - case, so any utility worth its salt will copy the MBR.

*TimDaniels*

 

[Bob Davis]

They're wrong.

I tend to agree, because as I said I've had the clone active for hours
without the slightest issue. The idea apparently is that NT/2K/XP can get
identical folders and files confused on the two drives, which won't cause a
problem if that drive stays on line continually. This sounds specious to
me, as how could the OS confuse two different drive letters? Anyway, in the
interest of safety I still try to keep the clone detached when possible.
However, see quotes at the end of this message.

You wont as long as you dont boot it.

Don't boot with what? The clone? If you clone a new drive and then use it
as the replacement for C:, you've just booted from the clone, and I've done
this many times. If you mean boot with the clone with the original drive
also on line, that is outside my experience, but I can't imagine ever doing
it.

When I've done it in the past I get a "configuring new hardware" window, but
that's it.

True Image is much better and you dont have
to boot from floppy or CD if you dont want to.

Perhaps, but cloning in Ghost 2003 has become a ritual for years, and it has
never failed me. If it ain't broke.... Also, I like to have a bonding
experience occasionally with DOS for old time's sake. I do it every Sat.
morning and not doing so would prolly confuse me for the rest of the day.

His approach gets it booted off the D drive much more
quickly. The only thing he has to do is unplug or depower
the C drive for the first boot of the D drive. He can plug
it back in again or repower it after XP has booted and
rebooted once and can get any files that have changed
since the last backup off the C drive in that config too
if the C drive isnt too bad and still spins up and mounts.

It would be easier, as I experienced for years with Win98SE, but I still
would rather be safe in dealing with XP. M$, Symantec, and Radified Ghost
(http://ghost.radified.com/ghost_1.htm) have warned against this practice.
I won't say it can't be done, but the spectre of a problem is enough to keep
me away. Here are two quotes from users (Radified Ghost, p. 13):

--------- Quote #1:

Your section on Cloning makes no mention of removing the newly created drive
from the system. Failure to do so before rebooting will annihilate your
registry.

We were moving an OS to a second drive. When cloning, you must remove the
cloned drive before rebooting into Windows. Windows will look at the system,
scan the registry, realizes its duplicated and deems it's corrupt. Then it
creates a new, blank registry, and carries on with that. I tried restoring
the registry from the command prompt, but alas nothing. Live and learn.

--------- Quote #2:

As I understand this issue, Windows XP "knows" which hardware was installed
when it is shut down. XP has attached a volume identifier to each volume.
When XP is restarted, it redetects the hardware and if the same, all is
well.

When a disk is cloned, disk-to-disk, there will be two volumes with the same
volume identifier. If the computer is restarted with both harddisks (the
"source" and the "clone") installed, XP will start from the "source", detect
the "clone" as new hardware and change the volume identifier since there can
not be two volumes with the same volume identifier.

Nothing will be detected by the user until he/she takes out the "source" and
makes the "clone" the boot drive. Now, XP can not boot because of the
changed volume identifier.

The solution is simple when Ghost 2003 is used to do the disk-to-disk
cloning. You don't let Ghost reset the computer and restart Windows but turn
off the computer and remove the "clone" before restarting. How this is done
using Ghost 9 I don't know.


---------- Quote #3 from Symantec (http://tinyurl.com/gh35f):

CAUTION: Do not start the computer after cloning until the instructions say
to do so. Starting a computer from the hard drive when the computer has two
Active partitions can damage program installations and trigger configuration
changes that you might not be able to reverse without restoring backups.

---------- End of quotes

 

[Bob Davis]

When Windows sees two identical drives, it changes the disk signature
of the second one. So if you want to be able to boot the clone, save a
copy of the orignal drive's MBR. When it becomes time to boot the
clone, replace the clone's MBR with the one you saved. Then you'll be
able to boot the clone.

Interesting. How do you do that?

 

[Timothy Daniels]

Back up Drive C to Drive D so that if something gets
screwed up i can simply remove Drive C and then
the "D" drive would become an exact replacement
which would pick up where "C" left off.

Is this possible?

[..............]

If thiis is indeed possible....how can a backup drive which
is a clone of the original boot at the same time as the
system drive, how does windows know the difference?

The hard drive at the head of the BIOS's hard drive boot
order is selected to control the booting. In the DEFAULT case
for most systems (those that whose hard drive boot order
cannot or has not been reset), this order is:
Master, IDE ch. 0,
Slave, IDE ch. 0,
Master, IDE ch. 1,
Slave, IDE ch. 1.

When you remove a hard drive from the head (top) of
the list, the next one in order is selected for booting. Since
most users connect the drive containing their operating
system as Master on IDE ch. 0, that one gets control at
boot time. Since the operating system is usually installed
in an isolated drive with no other partitions at the time, the
OS names its own partition "Local Disk (C", and it will
know other partitions that come to its attention as "Local
Disk (D", etc. Thus, when you boot while 2 hard drives
are connected, the one set as Master on ch. 0 (where you
have your "C" drive) will start booting. When you remove
that drive, the next one, the one set as Slave on ch. 0 - your
"D" drive - will start booting.

The above is a crude form of multi-booting, but it will
work as long as the partition containing the boot files
(usually the same one that has the OS) is a Primary partition
and it's marked "active". Cloning utilies will automatically
mark the clone's partition "active" or it will offer it as an option,
and there may be a similar option for putting the clone in a
Primary partition or an Extended partition.

*TimDaniels*

 

[Timothy Daniels]

"Bob Davis" mused:

Interesting. How do you do that?

*TimDaniels*

 

[Timothy Daniels]

The idea apparently is that NT/2K/XP can get
identical folders and files confused on the two drives,
which won't cause a problem if that drive stays on line
continually. This sounds specious to me, as how
could the OS confuse two different drive letters? Anyway,
in the interest of safety I still try to keep the clone detached
when possible.

The clone only has to be isolated from its "parent" OS
when it's started for its first time. How it will confuse itself
is quite mysterious, but i"ve seen it happen. Somehow,
addresses of random files in the clone point not to files
in the clone but to the correspondingly-named files in the
"parent" OS. I spotted this happening in My Documents.
You can go on editing and updating those documents in
what you think is the clone, but they're really being edited
and updated in the "parent". When you take the "parent"
away, the documents suddenly are inaccessible in the
clone. This is very bad for archiving and system backup,
as you can imagine.

But once the clone has been started up without its
"parent" visible, it becomes an independent OS, able to
live on its own - an "adult OS", if you will - and subsequently
it can be started up with its "parent" visible to it without it
forming any such linkages. The "parent" OS's partition
appears as just another partition having a different
letter assigned as its name - usually "D", so the "parent's"
partition is known to the clone as "Locasl Disk (D".

This startup isolation is not necessary for the "parent",
though - the "parent" can be started up numerous times
with its still unstarted clone entirely visible to it, and no
linkages will be made between the two OSes. You can
use this to your advantage to set the "active" flag in the
clone's partition, to readjust entries in the clone's boot.ini
file, to put a unique folder on the clone's Desktop to make
it easily identifieable at startup, etc. - all before starting
the clone for its first time. It's just necessary that the CLONE
be started for its FIRST TIME in ISOLATION from its "parent".

*TimDaniels*

 

[Rod Speed]

I tend to agree, because as I said I've had the clone active for hours
without the slightest issue.

And I've had it active for a hell of a lot longer than that.

The idea apparently is that NT/2K/XP can get identical folders and files
confused on the two drives,

Mindlessly silly.

which won't cause a problem if that drive stays on line continually.

Makes no sense either.

This sounds specious to me, as how could the OS confuse two different
drive letters?

Maybe they were suggesting that the USER might get confused.

Anyway, in the interest of safety I still try to keep the clone detached
when possible.

That is a bit safer if you want to protect yourself
against the very unlikely possibility of the power
supply dying and frying both drives at once.

I prefer to protect against that possibility by having
the drives on different PCs instead, and dont clone
that much, I find that images have significant advantages,
particularly with the new incremental images.

However, see quotes at the end of this message.

Don't boot with what? The clone?
Yep.

If you clone a new drive and then use it as the replacement for C:,
you've just booted from the clone, and I've done this many times. If you
mean boot with the clone with the original drive also on line,

Thats clearly the config you were talking about just above.

that is outside my experience, but I can't imagine ever doing it.

Quite a few do, basically to try the clone after
creating it to be sure that it can be booted.

That does get NT/2K/XP massively confused. If you want to try the
clone, its crucial to not allow it to see the original for the first boot
of the clone and the reboot of that clone which it will ask for.

When I've done it in the past I get a "configuring new hardware"
window, but that's it.

Yep, if you boot the clone without the original being visible,
it will claim to have found new hardware and will ask to be
allowed to reboot. Once you have rebooted, you can make
the original visible to the system again and you will be able
to boot either of the copys with impunity.

Perhaps,

No perhaps about it.

but cloning in Ghost 2003 has become a ritual for years, and it has never
failed me.

Its significantly crippled in capability.

If it ain't broke....

It is significantly crippled in capability.

Also, I like to have a bonding experience occasionally with DOS for old
time's sake.

More fool you. Makes much more sense to
consign it to the bin where it belongs now.

I do it every Sat. morning and not doing so would prolly confuse me for
the rest of the day.

Your problems with OCD are your problem.

It would be easier, as I experienced for years with Win98SE, but I still
would rather be safe in dealing with XP.

The only time it isnt safe is if you arent reliable enough to
unplug or depower the C drive for the first boot of the D drive.

M$, Symantec, and Radified Ghost (http://ghost.radified.com/ghost_1.htm)
have warned against this practice.

I dont care, I have tested the question thoroughly, and did that
because what they said didnt actually fit the available evidence.

I won't say it can't be done, but the spectre
of a problem is enough to keep me away.

Thats not what careful testing is about. It aint black magic
and you dont have to hold your mouth just right for it to work.
You just have to understand the basics and how to test properly.

Here are two quotes from users (Radified Ghost, p. 13):

Thats a steaming turd that has heaps of terminal stupiditys.

--------- Quote #1:

Your section on Cloning makes no mention of removing the newly created
drive from the system. Failure to do so before rebooting will annihilate
your registry.

Pig ignorant silly stuff.

We were moving an OS to a second drive. When cloning, you must remove the
cloned drive before rebooting into Windows.

Thats saying the exact opposite of the
one above on what needs to be removed.

Windows will look at the system, scan the registry, realizes its
duplicated and deems it's corrupt.
Wrong.

Then it creates a new, blank registry, and carries on with that.
Wrong.

I tried restoring the registry from the command prompt, but alas nothing.

Its quite possible to recover from that manually too.

Live and learn.

Nothing viable between its ears to learn with.

--------- Quote #2:

As I understand this issue, Windows XP "knows" which hardware was
installed when it is shut down.

Its MUCH more complicated than that too.

XP has attached a volume identifier to each volume. When XP is restarted,
it redetects the hardware and if the same, all is well.

When a disk is cloned, disk-to-disk, there will be two volumes with the
same volume identifier. If the computer is restarted with both harddisks
(the "source" and the "clone") installed, XP will start from the
"source", detect the "clone" as new hardware and change the volume
identifier since there can not be two volumes with the same volume
identifier.

No big deal.

Nothing will be detected by the user until he/she takes out the "source"
and makes the "clone" the boot drive. Now, XP can not boot because of the
changed volume identifier.

Wrong. And you have proved that is wrong and anyone can prove its wrong.

The solution is simple when Ghost 2003 is used to do the disk-to-disk
cloning. You don't let Ghost reset the computer and restart Windows
but turn off the computer and remove the "clone" before restarting.

Not necessary, as you have found.

How this is done using Ghost 9 I don't know.

Not even possible with 9 and 10 because they can only clone
when the OS is booted, they cant clone from the booted CD.

---------- Quote #3 from Symantec (http://tinyurl.com/gh35f):

CAUTION: Do not start the computer after cloning until the
instructions say to do so. Starting a computer from the hard drive
when the computer has two Active partitions can damage program
installations and trigger configuration changes that you might not be
able to reverse without restoring backups.

Just plain wrong. The ONLY thing that you should
avoid is booting the clone with the original visible.

And its completely trivial to test this and prove it.

 

[Rod Speed]

Any utility that will clone a drive will also copy the MBR - either
automatically or as a user-indicated option. Wanting a clone
that doesn't have an MBR that will boot it is an exceptional - even
pathological - case, so any utility worth its salt will copy the MBR.

Not necessarily, particularly with those which can
clone just a partition, not the entire physical drive.

 

[Rod Speed]

Interesting. How do you do that?

Just about any disk editor can do that and there
are specific MBR save and restore utes too.

 

[Rod Speed]

GTS wrote

The hard drive at the head of the BIOS's hard drive boot order is
selected to control the booting.

Its more complicated than that, most obviously
when it isnt bootable for whatever reason.

In the DEFAULT case for most systems (those that whose hard drive boot
order cannot or has not been reset), this order is:
Master, IDE ch. 0,
Slave, IDE ch. 0,
Master, IDE ch. 1,
Slave, IDE ch. 1.

Thats just plain wrong too. Heaps of systems dont
default any order like that and just allow you to
specify which of C cdrom or floppy to boot.

When you remove a hard drive from the head (top) of
the list, the next one in order is selected for booting.

Wrong again.

Since most users connect the drive containing their operating
system as Master on IDE ch. 0, that one gets control at
boot time. Since the operating system is usually installed
in an isolated drive with no other partitions at the time, the OS names
its own partition "Local Disk (C",

Utterly mangled all over again.

and it will know other partitions that come to its attention as "Local
Disk (D", etc.

That will happen even if you deliberately delete the OS partition
on a drive that has other partitions on it and reinstall the OS too.

Thus, when you boot while 2 hard drives are connected, the one set as
Master on ch. 0 (where you have your "C" drive) will start booting. When
you remove that drive, the next one, the one set as Slave on ch. 0 - your
"D" drive - will start booting.

Not necessarily, depends on the bios.

The above is a crude form of multi-booting, but it will
work as long as the partition containing the boot files
(usually the same one that has the OS) is a Primary partition
and it's marked "active". Cloning utilies will automatically
mark the clone's partition "active" or it will offer it as an option, and
there may be a similar option for putting the clone in a Primary
partition or an Extended partition.

That last is utterly mangled all over again.