"Not enough space for User Environment" Error

  • Thread starter Thread starter Nigel Walker
  • Start date Start date
N

Nigel Walker

I did the following steps:

Put single 500 Meg partition on my hard disk drive.
Built a working version of XPe on the hard disk.
Installed my application including some drivers for my hardware.
Ran fbreseal and stopped it at the point where it was going to boot up
again.
Copied this image to another computer.
Copied this image to a freshly formattted 512 MByte compact flash card.
Ran Bootprep.

It looked like it was booting fine and then I got the message "Not enough
space for User Environment". Then all I get is a static screen with the
Windows XP logo on.

Is this something to do with a slight difference in size between the disk I
used to build the system and the compact flash? Should I make one a lot
bigger or smaller than the other? Or is there something I haven't done?

Nigel.
 
I've done some more tests as follows:

Created a 450 Meg partition on the original hard disk.
Copied my fbresealed image to it.
Ran Bootprep.

I was expecting this to fail like the CF had done, but it didn't, it booted
up with no problems.

I'm confused now!

Nigel.
 
Nigel,

What are your original computer and target computer storage hardware design? I mean what HDDs, CFs and partitions on them?
Try deleting [HKLM\SYSTEM\MountedDevices] key jsut after you reseal the image and ready to shutdown the system. (or do that in the
system reg.hive offline)

Also, since you mentioned it hanged on Logo screen, can you add /SOS switch to boot.ini to see if drivers are loaded fine?

KM
 
Hi KM,

My development machine just has one big disk with one partition.

I build the system and allow it to boot on a hard disk drive connected to
IDE0 Master on the target board. Subsequently, I transfer the system to
compact flash and change the bios so that it will boot from IDE1 Master
(which has the compact flash in it). I suspect this is where the problem
lies.

I was assuming that Windows always talks to the BIOS to find/access the
disks. Therefore it would always expect to boot from drive C which is what I
had arranged. If this is not the case, do I need to change a registry entry
to point to the right physical location (IDE1 instead of IDE0)? If so, can I
set up the appropriate component in TD to deal with this or do I have to
patch the registry each time I build the system?

Regards,
Nigel.



KM said:
Nigel,

What are your original computer and target computer storage hardware
design? I mean what HDDs, CFs and partitions on them?
Try deleting [HKLM\SYSTEM\MountedDevices] key jsut after you reseal the
image and ready to shutdown the system. (or do that in the system reg.hive
offline)

Also, since you mentioned it hanged on Logo screen, can you add /SOS
switch to boot.ini to see if drivers are loaded fine?

KM
I've done some more tests as follows:

Created a 450 Meg partition on the original hard disk.
Copied my fbresealed image to it.
Ran Bootprep.

I was expecting this to fail like the CF had done, but it didn't, it
booted up with no problems.

I'm confused now!

Nigel.
 
Nigel,

Did you disconnect HDD when you transferred the image to CF? That was not clear from your message.
Please do that and clean up the reg.key I mentioned earlier just before the transfer.

No, Windows does not really rely on BIOS to assign drive letter.
Read this tip to get a clear picture on the driver letter assignment:
http://msdn.microsoft.com/embedded/community/community/tips/xp/rtpartin/default.aspx

--
Regards,
KM, BSquare Corp.

Hi KM,

My development machine just has one big disk with one partition.

I build the system and allow it to boot on a hard disk drive connected to
IDE0 Master on the target board. Subsequently, I transfer the system to
compact flash and change the bios so that it will boot from IDE1 Master
(which has the compact flash in it). I suspect this is where the problem
lies.

I was assuming that Windows always talks to the BIOS to find/access the
disks. Therefore it would always expect to boot from drive C which is what I
had arranged. If this is not the case, do I need to change a registry entry
to point to the right physical location (IDE1 instead of IDE0)? If so, can I
set up the appropriate component in TD to deal with this or do I have to
patch the registry each time I build the system?

Regards,
Nigel.



KM said:
Nigel,

What are your original computer and target computer storage hardware
design? I mean what HDDs, CFs and partitions on them?
Try deleting [HKLM\SYSTEM\MountedDevices] key jsut after you reseal the
image and ready to shutdown the system. (or do that in the system reg.hive
offline)

Also, since you mentioned it hanged on Logo screen, can you add /SOS
switch to boot.ini to see if drivers are loaded fine?

KM
I've done some more tests as follows:

Created a 450 Meg partition on the original hard disk.
Copied my fbresealed image to it.
Ran Bootprep.

I was expecting this to fail like the CF had done, but it didn't, it
booted up with no problems.

I'm confused now!

Nigel.

I did the following steps:

Put single 500 Meg partition on my hard disk drive.
Built a working version of XPe on the hard disk.
Installed my application including some drivers for my hardware.
Ran fbreseal and stopped it at the point where it was going to boot up
again.
Copied this image to another computer.
Copied this image to a freshly formattted 512 MByte compact flash card.
Ran Bootprep.

It looked like it was booting fine and then I got the message "Not
enough space for User Environment". Then all I get is a static screen
with the Windows XP logo on.

Is this something to do with a slight difference in size between the
disk I used to build the system and the compact flash? Should I make one
a lot bigger or smaller than the other? Or is there something I haven't
done?

Nigel.
 
Ah, I forgot to mention that now that info is a part of XPE SP2 docs:
http://msdn.microsoft.com/library/e...signDriveLettersToVolumesInTargetDesigner.asp

--
Regards,
KM, BSquare Corp.

Nigel,

Did you disconnect HDD when you transferred the image to CF? That was not clear from your message.
Please do that and clean up the reg.key I mentioned earlier just before the transfer.

No, Windows does not really rely on BIOS to assign drive letter.
Read this tip to get a clear picture on the driver letter assignment:
http://msdn.microsoft.com/embedded/community/community/tips/xp/rtpartin/default.aspx

--
Regards,
KM, BSquare Corp.

Hi KM,

My development machine just has one big disk with one partition.

I build the system and allow it to boot on a hard disk drive connected to
IDE0 Master on the target board. Subsequently, I transfer the system to
compact flash and change the bios so that it will boot from IDE1 Master
(which has the compact flash in it). I suspect this is where the problem
lies.

I was assuming that Windows always talks to the BIOS to find/access the
disks. Therefore it would always expect to boot from drive C which is what I
had arranged. If this is not the case, do I need to change a registry entry
to point to the right physical location (IDE1 instead of IDE0)? If so, can I
set up the appropriate component in TD to deal with this or do I have to
patch the registry each time I build the system?

Regards,
Nigel.



KM said:
Nigel,

What are your original computer and target computer storage hardware
design? I mean what HDDs, CFs and partitions on them?
Try deleting [HKLM\SYSTEM\MountedDevices] key jsut after you reseal the
image and ready to shutdown the system. (or do that in the system reg.hive
offline)

Also, since you mentioned it hanged on Logo screen, can you add /SOS
switch to boot.ini to see if drivers are loaded fine?

KM

I've done some more tests as follows:

Created a 450 Meg partition on the original hard disk.
Copied my fbresealed image to it.
Ran Bootprep.

I was expecting this to fail like the CF had done, but it didn't, it
booted up with no problems.

I'm confused now!

Nigel.

I did the following steps:

Put single 500 Meg partition on my hard disk drive.
Built a working version of XPe on the hard disk.
Installed my application including some drivers for my hardware.
Ran fbreseal and stopped it at the point where it was going to boot up
again.
Copied this image to another computer.
Copied this image to a freshly formattted 512 MByte compact flash card.
Ran Bootprep.

It looked like it was booting fine and then I got the message "Not
enough space for User Environment". Then all I get is a static screen
with the Windows XP logo on.

Is this something to do with a slight difference in size between the
disk I used to build the system and the compact flash? Should I make one
a lot bigger or smaller than the other? Or is there something I haven't
done?

Nigel.
 
Those are great links KM, it gives just the insight I was after on what goes
on, many thanks.

I have sidestepped the issue a bit now by putting a Microdrive into the CF
slot with a system on it, letting that boot up and then fbreseal on that. I
then use that as the source for my target image for transferring onto real
compact flash and that, as expected, works fine.

One question, when I do a fbreseal and I get the "resealed" ok box, do I
allow the machine close itself down or do I kill the machine right there? Or
doesn't it matter?

Many thanks for your help, KM.

Regards,
Nigel.

KM said:
Ah, I forgot to mention that now that info is a part of XPE SP2 docs:
http://msdn.microsoft.com/library/e...signDriveLettersToVolumesInTargetDesigner.asp

--
Regards,
KM, BSquare Corp.

Nigel,

Did you disconnect HDD when you transferred the image to CF? That was not
clear from your message.
Please do that and clean up the reg.key I mentioned earlier just before
the transfer.

No, Windows does not really rely on BIOS to assign drive letter.
Read this tip to get a clear picture on the driver letter assignment:
http://msdn.microsoft.com/embedded/community/community/tips/xp/rtpartin/default.aspx

--
Regards,
KM, BSquare Corp.

Hi KM,

My development machine just has one big disk with one partition.

I build the system and allow it to boot on a hard disk drive connected
to
IDE0 Master on the target board. Subsequently, I transfer the system to
compact flash and change the bios so that it will boot from IDE1 Master
(which has the compact flash in it). I suspect this is where the
problem
lies.

I was assuming that Windows always talks to the BIOS to find/access the
disks. Therefore it would always expect to boot from drive C which is
what I
had arranged. If this is not the case, do I need to change a registry
entry
to point to the right physical location (IDE1 instead of IDE0)? If so,
can I
set up the appropriate component in TD to deal with this or do I have
to
patch the registry each time I build the system?

Regards,
Nigel.



Nigel,

What are your original computer and target computer storage hardware
design? I mean what HDDs, CFs and partitions on them?
Try deleting [HKLM\SYSTEM\MountedDevices] key jsut after you reseal
the
image and ready to shutdown the system. (or do that in the system
reg.hive
offline)

Also, since you mentioned it hanged on Logo screen, can you add /SOS
switch to boot.ini to see if drivers are loaded fine?

KM

I've done some more tests as follows:

Created a 450 Meg partition on the original hard disk.
Copied my fbresealed image to it.
Ran Bootprep.

I was expecting this to fail like the CF had done, but it didn't, it
booted up with no problems.

I'm confused now!

Nigel.

I did the following steps:

Put single 500 Meg partition on my hard disk drive.
Built a working version of XPe on the hard disk.
Installed my application including some drivers for my hardware.
Ran fbreseal and stopped it at the point where it was going to boot
up
again.
Copied this image to another computer.
Copied this image to a freshly formattted 512 MByte compact flash
card.
Ran Bootprep.

It looked like it was booting fine and then I got the message "Not
enough space for User Environment". Then all I get is a static
screen
with the Windows XP logo on.

Is this something to do with a slight difference in size between
the
disk I used to build the system and the compact flash? Should I
make one
a lot bigger or smaller than the other? Or is there something I
haven't
done?

Nigel.
 
Nigel,

It is always preferable to shutdown Windows properly.

After FBA reseal you do want to shutdown it properly (e.g., to flash all the changes).

The cloning phase will happen on the next on any device you copied the image to.

--
Regards,
KM, BSquare Corp.

Those are great links KM, it gives just the insight I was after on what goes
on, many thanks.

I have sidestepped the issue a bit now by putting a Microdrive into the CF
slot with a system on it, letting that boot up and then fbreseal on that. I
then use that as the source for my target image for transferring onto real
compact flash and that, as expected, works fine.

One question, when I do a fbreseal and I get the "resealed" ok box, do I
allow the machine close itself down or do I kill the machine right there? Or
doesn't it matter?

Many thanks for your help, KM.

Regards,
Nigel.

KM said:
Ah, I forgot to mention that now that info is a part of XPE SP2 docs:
http://msdn.microsoft.com/library/e...signDriveLettersToVolumesInTargetDesigner.asp

--
Regards,
KM, BSquare Corp.

Nigel,

Did you disconnect HDD when you transferred the image to CF? That was not
clear from your message.
Please do that and clean up the reg.key I mentioned earlier just before
the transfer.

No, Windows does not really rely on BIOS to assign drive letter.
Read this tip to get a clear picture on the driver letter assignment:
http://msdn.microsoft.com/embedded/community/community/tips/xp/rtpartin/default.aspx

--
Regards,
KM, BSquare Corp.


Hi KM,

My development machine just has one big disk with one partition.

I build the system and allow it to boot on a hard disk drive connected
to
IDE0 Master on the target board. Subsequently, I transfer the system to
compact flash and change the bios so that it will boot from IDE1 Master
(which has the compact flash in it). I suspect this is where the
problem
lies.

I was assuming that Windows always talks to the BIOS to find/access the
disks. Therefore it would always expect to boot from drive C which is
what I
had arranged. If this is not the case, do I need to change a registry
entry
to point to the right physical location (IDE1 instead of IDE0)? If so,
can I
set up the appropriate component in TD to deal with this or do I have
to
patch the registry each time I build the system?

Regards,
Nigel.



Nigel,

What are your original computer and target computer storage hardware
design? I mean what HDDs, CFs and partitions on them?
Try deleting [HKLM\SYSTEM\MountedDevices] key jsut after you reseal
the
image and ready to shutdown the system. (or do that in the system
reg.hive
offline)

Also, since you mentioned it hanged on Logo screen, can you add /SOS
switch to boot.ini to see if drivers are loaded fine?

KM

I've done some more tests as follows:

Created a 450 Meg partition on the original hard disk.
Copied my fbresealed image to it.
Ran Bootprep.

I was expecting this to fail like the CF had done, but it didn't, it
booted up with no problems.

I'm confused now!

Nigel.

I did the following steps:

Put single 500 Meg partition on my hard disk drive.
Built a working version of XPe on the hard disk.
Installed my application including some drivers for my hardware.
Ran fbreseal and stopped it at the point where it was going to boot
up
again.
Copied this image to another computer.
Copied this image to a freshly formattted 512 MByte compact flash
card.
Ran Bootprep.

It looked like it was booting fine and then I got the message "Not
enough space for User Environment". Then all I get is a static
screen
with the Windows XP logo on.

Is this something to do with a slight difference in size between
the
disk I used to build the system and the compact flash? Should I
make one
a lot bigger or smaller than the other? Or is there something I
haven't
done?

Nigel.
 
Back
Top