How Partition Numbers are Assigned in BOOT.INI

[Larry(LJL269)]

http://www.bootitng.com/kb/article.php?id=159 nor
anything else I've read mentions whether to count
hidden Partitions. I dual boot XP so I have to hide a
primary with PQboot or PM7 but I'm gonna try both
unhidden with 1 active.

PS- in my case, only 1 partition entry works fine & the
OS selections NEVER appear!

Your help is MUCH appreciated. Thanks- bye- Larry

Any advise is my attempt to contribute more than I have received but I can only assure you that it works on my PC. GOOD LUCK.

 

[Doug Knox MS-MVP]

I can't speak to PQBoot, but for BootIt NG, it uses a method of faking out the BIOS. Even though a partition may be the 3rd physical partition on the drive, you can make the computer believe its the first partition on the drive, by use of the Boot Menu items.

 

[Larry(LJL269)]

Greetings & thank you for your response.

In my case, I have 2 primaries at start of HD followed by a
Logical. PQboot just hides primary u dont want to boot &
sets active the 1 u do. U can do same thing in PM7- just
takes longer. XP will boot from 1st Active primary whether
its 1st or 2nd partition & whether its preceeded by an
unhidden primary.

I had lots problems unhiding other XP that wasnt active- XP
kept treating other partition as an OS, not data. Might this
be.solved by unhiding 2nd XP thats not active but having NO
entry in BOOT.INI for it? ie:here's my BOOT.INI now
[boot loader]
timeout=5
default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS
[operating systems]
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="XPmin @ Part1
" /fastdetect /bootlog /NoExecute=OptOut
What happens if I unhide partition(2) & leave BOOT.INI
alone?

All this is to see if I can get XP1 to repair XP2 like
BartPE does. Theoretically it's same thing.

Gonna post comparison of BartPE vs Recovery Console. 1st
time I've used either. BartPE appears to be way better than
RC.

Comments/suggestions/corrections appreciated.
Thanks- bye- Larry

I can't speak to PQBoot, but for BootIt NG, it uses a method of faking out the BIOS. Even though a partition may be the 3rd physical partition on the drive, you can make the computer believe its the first partition on the drive, by use of the Boot Menu items.

Any advise given is my attempt to show appreciation for all
the excellent help I've received here but I'm no MVP so it
may only apply NUGS (Normally, Usually, Generally, Sometimes

 

[Ron Sommer]

The boot.ini on the second primary partition has to be:

[boot loader]
timeout=5
default=multi(0)disk(0)rdisk(0)partition(2)\WINDOWS
[operating systems]
multi(0)disk(0)rdisk(0)partition(2)\WINDOWS="XPmin @ Part1
" /fastdetect /bootlog /NoExecute=OptOut
--
Ron Sommer

Greetings & thank you for your response.

In my case, I have 2 primaries at start of HD followed by a
Logical. PQboot just hides primary u dont want to boot &
sets active the 1 u do. U can do same thing in PM7- just
takes longer. XP will boot from 1st Active primary whether
its 1st or 2nd partition & whether its preceeded by an
unhidden primary.

I had lots problems unhiding other XP that wasnt active- XP
kept treating other partition as an OS, not data. Might this
be.solved by unhiding 2nd XP thats not active but having NO
entry in BOOT.INI for it? ie:here's my BOOT.INI now
[boot loader]
timeout=5
default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS
[operating systems]
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="XPmin @ Part1
" /fastdetect /bootlog /NoExecute=OptOut
What happens if I unhide partition(2) & leave BOOT.INI
alone?

All this is to see if I can get XP1 to repair XP2 like
BartPE does. Theoretically it's same thing.

Gonna post comparison of BartPE vs Recovery Console. 1st
time I've used either. BartPE appears to be way better than
RC.

Comments/suggestions/corrections appreciated.
Thanks- bye- Larry

I can't speak to PQBoot, but for BootIt NG, it uses a method of faking out
the BIOS. Even though a partition may be the 3rd physical partition on
the drive, you can make the computer believe its the first partition on
the drive, by use of the Boot Menu items.

Any advise given is my attempt to show appreciation for all
the excellent help I've received here but I'm no MVP so it
may only apply NUGS (Normally, Usually, Generally, Sometimes

 

[Larry(LJL269)]

Greetings Ron & thank you for your response.

I've been booting with (1) not (2) for 2 yrs now. My
conjecture is u can put any number in there & it will boot
the only active primary in dual boot XP like mine. I may
experiment.

So the fact it boots the correct partition in spite of an
incorrect boot.ini makes me wonder what its purpose is.

Comments/suggestions/corrections appreciated.
Thanks- bye- Larry

The boot.ini on the second primary partition has to be:

[boot loader]
timeout=5
default=multi(0)disk(0)rdisk(0)partition(2)\WINDOWS
[operating systems]
multi(0)disk(0)rdisk(0)partition(2)\WINDOWS="XPmin @ Part1
" /fastdetect /bootlog /NoExecute=OptOut

Any advise given is my attempt to show appreciation for all
the excellent help I've received here but I'm no MVP so it
may only apply NUGS (Normally, Usually, Generally, Sometimes

 

[Ron Sommer]

You have two boot.ini files.
It is booting the active partition and the partition number in partition (1)
is correct in the partition (1) boot.ini.
Are you saying that both boot.ini files have (1)?
--
Ron Sommer

Greetings Ron & thank you for your response.

I've been booting with (1) not (2) for 2 yrs now. My
conjecture is u can put any number in there & it will boot
the only active primary in dual boot XP like mine. I may
experiment.

So the fact it boots the correct partition in spite of an
incorrect boot.ini makes me wonder what its purpose is.

Comments/suggestions/corrections appreciated.
Thanks- bye- Larry

The boot.ini on the second primary partition has to be:

[boot loader]
timeout=5
default=multi(0)disk(0)rdisk(0)partition(2)\WINDOWS
[operating systems]
multi(0)disk(0)rdisk(0)partition(2)\WINDOWS="XPmin @ Part1
" /fastdetect /bootlog /NoExecute=OptOut

Any advise given is my attempt to show appreciation for all
the excellent help I've received here but I'm no MVP so it
may only apply NUGS (Normally, Usually, Generally, Sometimes

 

[Larry(LJL269)]

Hi Ron

I have a HD with p1 p2 L where p1,p2 = primaries & L=logical
I am running XP on p2(assigned to C, p1 is hidden
Only boot.ini in C: contains :
[boot loader]
timeout=5
default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS
[operating systems]
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="XPmin @ Part1
" /fastdetect /bootlog /NoExecute=OptOut

which is wrong cause (1) is p1which is Hidden.

Moreover, until 4/23/06 this boot.ini worked & was changed
to produce the 1 above:
[boot loader]
timeout=3
default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS
[operating systems]
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="72-1-XP MIN X
OS" /fastdetect /bootlog /NoExecute=OptOut
multi(0)disk(0)rdisk(1)partition(1)\WINDOWS="72-0-XP MAX
(#1)" /fastdetect /bootlog

On p1, which has XP also, I have ~3 bootini's that all
worked. I always save old bootini's as bootmmddyy. When I
switch to other XP on p1 I can list them if u like.

So I gotta conclude bootini may have to exist but in 3
configurations I boot to its contents is immaterial. So the
question is: What other functions does it serve?

As a scientist, contradictions fuel curiosity.

Comments/suggestions/corrections appreciated.
Thanks- bye- Larry

You have two boot.ini files.
It is booting the active partition and the partition number in partition (1)
is correct in the partition (1) boot.ini.
Are you saying that both boot.ini files have (1)?

Any advise given is my attempt to show appreciation for all
the excellent help I've received here but I'm no MVP so it
may only apply NUGS (Normally, Usually, Generally, Sometimes

 

[Timothy Daniels]

So I gotta conclude bootini may have to exist but in 3
configurations I boot to its contents is immaterial. So the
question is: What other functions does it serve?

Part of the problem is that you don't understand how
a boot.ini file gets control. In the boot process, the MB
(master boot record) of the HD that is at the head of the
BIOS's HD boot order gets control.

That MBR's executable code looks for the primary partition
that is marked "active" and passes control to it.

The boot sector of that "active" partition looks for ntldr and
passes control to it. ntldr looks for boot.ini and if there are
more than 1 entry under the line "[operating systems]", it
displays the textual portion of boot.ini's entries as a menu
on the screen. Otherwise, ntldr just passes control to the
entry listed as "default".

The other part of the problem (which you're aware of) is
how hiding a partition affects the meaning of "x" in the
"partition(x)" segment of each boot.ini entry. IOW, can
"partition(x)" refer to a partition that is hidden, or does
the numbering space close up to ignore the hidden
partition? Once you understand how a ntldr is selected
you'll know which boot.ini's contents are used for the
boot menu, and through experimentation you'll be able
to answer the 2nd question yourself. When you do,
please post to let us know, too.

BTW, a convenient way to know which OS has booted
is to put a file with an identifying name on the Desktop
and at the root of the file system. When the system runs,
the Desktop will show which OS it is. When it's not running,
and a peak at the root of each partition will how which OS
resides there.

*TimDaniels*

 

[cquirke (MVP Windows shell/user)]

On Sun, 7 May 2006 16:40:41 -0700, "Timothy Daniels"

Part of the problem is that you don't understand how
a boot.ini file gets control. In the boot process, the MBR
(master boot record) of the HD that is at the head of the
BIOS's HD boot order gets control.

That MBR's executable code looks for the primary partition
that is marked "active" and passes control to it.

That's where the "system" phase ends and the "OS" begins.

The boot sector of that "active" partition looks for ntldr and
passes control to it. ntldr looks for boot.ini and if there are
more than 1 entry under the line "[operating systems]", it
displays the textual portion of boot.ini's entries as a menu
on the screen. Otherwise, ntldr just passes control to the
entry listed as "default".

True, where the boot sector is for the NT family. Other partition
boot record code may do other things, e.g. DOS or Win9x will look for
and pass control to IO.SYS instead, and other OSs do other things.

The other part of the problem (which you're aware of) is
how hiding a partition affects the meaning of "x" in the
"partition(x)" segment of each boot.ini entry. IOW, can
"partition(x)" refer to a partition that is hidden, or does
the numbering space close up to ignore the hidden
partition?

AFAIK, the Boot.ini syntax numbering counts which of the 4 possible
partition table entries it is, without any reference to what those
entries are - it's purely a matter of position.

AFAIK, hiding a partition doesn't change any order of entries; it
merely sets the partition type byte to something the OS doesn't
understand, and is expected to ignore.

An OS that "discovers" (and delves into) partitions that are not
marked as of that OS's type is a badly-behaved OS looking for trouble.

------------ ----- ---- --- -- - - - -

The most accurate diagnostic instrument
in medicine is the Retrospectoscope

 

[Timothy Daniels]

The boot sector of that "active" partition looks for ntldr and
passes control to it. ntldr looks for boot.ini and if there are
more than 1 entry under the line "[operating systems]", it
displays the textual portion of boot.ini's entries as a menu
on the screen. Otherwise, ntldr just passes control to the
entry listed as "default".

True, where the boot sector is for the NT family.

And that is why I posted this in a Windows XP newsgroup.

AFAIK, the Boot.ini syntax numbering counts which of the
4 possible partition table entries it is, without any reference
to what those entries are - it's purely a matter of position.

The partition numbering in the boot.ini entries does follow
physical position, but that includes logical drives in an
extended partition, and the maximum number of logical
drives is "many". As an experiment, I have booted a
WinXP OS from the 2nd logical drive in an extended
partition, and Microsoft documentation only says that
the partition numbering *begins* with "1". It does not
mention a maximum no. as it does for "y" in "rdisk(y)".

*TimDaniels*







Thus, there could be a "partition(5)".
I assume you meant to say "it's purely a matter of position
on the controler's channels". But even that is not generally
true. I've done extensive experiments using my Dell PC's
Phoenix BIOS - one which allows adjustment of the hard
drive boot order (which I think is quite common nowadays) -
and the

 

[Larry(LJL269)]

Part of the problem is that you don't understand how
a boot.ini file gets control. In the boot process, the MB
(master boot record) of the HD that is at the head of the
BIOS's HD boot order gets control.

That MBR's executable code looks for the primary partition
that is marked "active" and passes control to it.

The boot sector of that "active" partition looks for ntldr and
passes control to it. ntldr looks for boot.ini and if there are
more than 1 entry under the line "[operating systems]", it
displays the textual portion of boot.ini's entries as a menu
on the screen. Otherwise, ntldr just passes control to the
entry listed as "default".

The other part of the problem (which you're aware of) is
how hiding a partition affects the meaning of "x" in the
"partition(x)" segment of each boot.ini entry. IOW, can
"partition(x)" refer to a partition that is hidden, or does
the numbering space close up to ignore the hidden
partition? Once you understand how a ntldr is selected
you'll know which boot.ini's contents are used for the
boot menu, and through experimentation you'll be able
to answer the 2nd question yourself. When you do,
please post to let us know, too.

BTW, a convenient way to know which OS has booted
is to put a file with an identifying name on the Desktop
and at the root of the file system. When the system runs,
the Desktop will show which OS it is. When it's not running,
and a peak at the root of each partition will how which OS
resides there.

Greetings Tim & thank you for your response.

TY for explaining how the boot process works. I tried to
Google lots of boot phrases but was inundated with
extraneous info.

Re 'convenient way to know which OS has booted
is to put a file with an identifying name on the Desktop
and at the root of the file system', I also have real
different wallpapers 4 each XP and each partitions has label
like
'B4-F DATA' (4th part on B HD contains data & should be F in
XP) & 1st folder is
'(B4-F DATA=Data=MyDocs + PgmData ) .

TY again-u were very helpfull

Any advise given is my attempt to show appreciation for all
the excellent help I've received here but I'm no MVP so it
may only apply NUGS (Normally, Usually, Generally, Sometimes

 

[cquirke (MVP Windows shell/user)]

:
True, where the boot sector is for the NT family.

[/QUOTE]

And that is why I posted this in a Windows XP newsgroup.

You were describing the process in a general way (or that was how I
was reading it) and that seemed to suggest all partition boot sectors
would loaf \NTLDR. Even in the world of XP, it's worth remembering
that a DOS/Win9x boot sector (e.g. side-effect from Sys C will
bypass \NTLDR and load IO.SYS directly.

The partition numbering in the boot.ini entries does follow
physical position, but that includes logical drives in an
extended partition, and the maximum number of logical
drives is "many".

That's interesting, and not at all what I'd have expected.

Logical volumes are not "partitions" in the system (i.e.
OS-independent) sense of the term. At this level, all the system
deals with is partitions as closed boxes with boot sectors in front,
and a marker to tell which one to boot. There's no knowledge of
whatever goes on in the box beyond that boot sector.

So I'd have expected the syntax to count system-level partitions,
irrespective of what type they might be.

Instead, you're implying it counts OS-visible volumes, which implies
knowledge of partition type and (in the case of extended partitions)
internal structure. If so, one might expect partitions not of the
OS's type - e.g. Linux, or partitions "hidden" by spoofing the type
byte - to be ignored in the count.

And if *that* is so, then I can see all kinds of mishap arising when
partitions are hidden and unhidden.

------------------------ ---- --- -- - - - -

On a bad day I'm tired and irritable.
On a good day I'm just tired.
Sorry if this is a bad day

 

[cquirke (MVP Windows shell/user)]

Re 'convenient way to know which OS has booted
is to put a file with an identifying name on the Desktop
and at the root of the file system', I also have real
different wallpapers 4 each XP

When it comes to multiple physical hard drives, a safer method is to
create a different fake do-nothing entry in the \Boot.ini of each.

That way, you can back out of booting XP if you realize you are about
to boot the wrong HD (e.g. one from another PC, before PnP and Product
Activation can *seriously* spoil your day)

---------- ----- ---- --- -- - - - -

Don't pay malware vendors - boycott Sony

 

[Larry(LJL269)]

When it comes to multiple physical hard drives, a safer method is to
create a different fake do-nothing entry in the \Boot.ini of each.

That way, you can back out of booting XP if you realize you are about
to boot the wrong HD (e.g. one from another PC, before PnP and Product
Activation can *seriously* spoil your day)

Greetings cquirke & thank you for your response.

Damn good idea but booting 2 XPs it seems to boot active
primary irregardless of bootini.

Any info re booting XP from 2nd HD with unhidden primaries
on 1st: HD0 xp1 xp2 L1,2,3
HD1 xp3
where xp(n)=primaries but only xp3 is ACTIVE &L(n) r
logicals
(Trying to make HD1 run like WPE )

Your help is MUCH appreciated. Thanks- bye- Larry


Any advise given is my attempt to show appreciation for all
the excellent help I've received here but I'm no MVP so it
may only apply NUGS (Normally, Usually, Generally, Sometimes

 

[cquirke (MVP Windows shell/user)]

Damn good idea but booting 2 XPs it seems to boot active
primary irregardless of bootini.

The problem arises when you mix SATA with IDE HDs, because BIOSs can
be unclear as to which will be booted first.

In fact, Intel's 1-year-old BIOS is still getting 1-2 updates a month
and is still shipping (as sourced 2 weeks ago) with significant logic
defects concerning boot order. What the BIOS menu settings say will
happen, is not necessarily what happens - VERY ugly indeed!

From one of Intel's .PDFs...

ftp://aiedownload.intel.com/df-support/10485/ENG/NT_3832_ReleaseNotes.pdf

April 11, 2006
.. NT94510J.86A.3813.2006.0411.1332
.. VBIOS info: Build Number: 1295

New Fixes/Features:
.. Fixed multiple TPM issues.
.. Fixed a problem where Boot Order in Setup will revert back to
default if the user "ESC"
from other sub-menu.
.. Fixed issue where legacy floppy drive was not seen in Windows if
"Boot to Removable
Devices" is set to "Disabled".

(with much snipped!)

Just what we need; "multiple TPM issues".

In practice, the new 945G Sorento has multiple "fuzzinesses" about the
boot order settings, such as being in Advanced mode when set to Normal
and vice versa, and booting "disabled" device categories before what
was left enabled even though the disabled items vanish from the boot
order list, etc. Quite surprisingly bad.

---------- ----- ---- --- -- - - - -

On the 'net, *everyone* can hear you scream

 

[Larry(LJL269)]

Greetings & thank you for your response.

I'm gonna image everything & try some combos of active,
hidden, & unhidden primaries on 2 HD's. I am confident using
PMagic & Image4windows boot disks & BIOS. Hardest part will
be keeping notes to produce intelligable results which I
shall post here (normally ignored in past as when I
explained all hotkey combos possable). At least the mental
exercise will keep me from getting alzheimers

Your help is MUCH appreciated.
Thanks- bye- Larry

The problem arises when you mix SATA with IDE HDs, because BIOSs can
be unclear as to which will be booted first.

In fact, Intel's 1-year-old BIOS is still getting 1-2 updates a month
and is still shipping (as sourced 2 weeks ago) with significant logic
defects concerning boot order. What the BIOS menu settings say will
happen, is not necessarily what happens - VERY ugly indeed!

From one of Intel's .PDFs...

ftp://aiedownload.intel.com/df-support/10485/ENG/NT_3832_ReleaseNotes.pdf

April 11, 2006
. NT94510J.86A.3813.2006.0411.1332
. VBIOS info: Build Number: 1295

New Fixes/Features:
. Fixed multiple TPM issues.
. Fixed a problem where Boot Order in Setup will revert back to
default if the user "ESC"
from other sub-menu.
. Fixed issue where legacy floppy drive was not seen in Windows if
"Boot to Removable
Devices" is set to "Disabled".

(with much snipped!)

Just what we need; "multiple TPM issues".

In practice, the new 945G Sorento has multiple "fuzzinesses" about the
boot order settings, such as being in Advanced mode when set to Normal
and vice versa, and booting "disabled" device categories before what
was left enabled even though the disabled items vanish from the boot
order list, etc. Quite surprisingly bad.



On the 'net, *everyone* can hear you scream

Any advise given is my attempt to show appreciation for all
the excellent help I've received here but I'm no MVP so it
may only apply NUGS (Normally, Usually, Generally, Sometimes