Hello dear guest!

Boot Land is a community driven pc software site established since 2006 and focused on recovery/backup boot disks, research of Microsoft Windows 2000/XP/2003/Vista/7 install/deployment/lease/antivirus/antispam tools, customizing Microsoft Windows PE administration systems and even learning how to recover computer data from disaster situations!

How about joining our boot disk community? So do it. Life's short!

  - You get free access to our newsletter with all the interesting buzz about boot disks
  - We share publicity revenue with everyone who wishes to participate at the forums
  - Publicity is never, never, never displayed to members (along with many other cool things)
http://boot-land.net/register


16 Pages V   1 2 3 > » 

dmio/ftdisk PE1 driver question
Sha0
Posted on: Today, 01:44 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (oriensol @ Jun 21 2010, 09:13 PM) *
I will test with ControlSet001\Enum\Root\dmio prepopulated in setupreg.hiv and report back after some time.

I doubt this will have any impact, since I do not believe that SETUPREG.HIV is used by SETUPLDR to determine which drivers to load before the kernel is invoked.
QUOTE (oriensol @ Jun 21 2010, 09:13 PM) *
To answer your question about txtsetup.sif, the only thing I find is:...

That looks completely sane... Unless you didn't also see dmio under the [BusExtenders] section...?

Could you be interested in running Microsoft's SysInternals' LoadOrd.Exe in the PE and having a look at the load order for the various drivers? It includes a copy-to-clipboard function too, in case you feel like sharing it here.
  Forum: Windows 2K/XP/2003 · Post Preview: #103103 · Replies: 25 · Views: 297

Switching to AHCI without BIOS support problem
Sha0
Posted on: Today, 12:57 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


Are you saying that you are changing device details after BIOS POST?! Are you changing any VENdor and DEVice IDs? Do those IDs have entries in the CriticalDeviceDatabase so that the driver is associated with the device?
  Forum: Boot from USB / Boot anywhere · Post Preview: #103099 · Replies: 3 · Views: 21

dmio/ftdisk PE1 driver question
Sha0
Posted on: Today, 12:54 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (oriensol @ Jun 21 2010, 08:20 PM) *
I have searched ALL the files - I don't see any file containing the text LEGACY_DMIO (or LEGACY_FTDISK), either in the livexp project/scripts or in the PE1 iso/target or in the (runtime) registry (other than the services\xxx\enum\legacy_xxx entry)...

What about just "dmio" without the LEGACY_ prefix? What lines show it up in LiveXP's TXTSETUP.SIF?
QUOTE (oriensol @ Jun 21 2010, 08:43 PM) *
...Shao says BartPE doesn't create this, there must be something else also, correct?...

Just to clarify, I tested before saying that.
  Forum: Windows 2K/XP/2003 · Post Preview: #103098 · Replies: 25 · Views: 297

Switching to AHCI without BIOS support problem
Sha0
Posted on: Today, 12:15 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


The Windows AHCI driver for your storage adapter needs to be loaded and invoked at boot time in order to avoid the 0x7B error. Attaching WinDbg, obtaining the CLI once the bug-check occurs, then entering !devnode 0 1 should reveal the status of each Windows device. Look for the storage adapter and find out what its trouble is. You might need to inject the driver into your OS installation, including CriticalDeviceDatabase entries for the corresponding PCI IDs.
  Forum: Boot from USB / Boot anywhere · Post Preview: #103092 · Replies: 3 · Views: 21

dmio/ftdisk PE1 driver question
Sha0
Posted on: Yesterday, 11:09 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


BartPE does not produce the LEGACY_xxx entry for dmio. How about looking for all lines with "dmio" in TXTSETUP.SIF?
  Forum: Windows 2K/XP/2003 · Post Preview: #103088 · Replies: 25 · Views: 297

ELTORITO.SYS and the CDROM error
Sha0
Posted on: Yesterday, 06:15 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


So just to confirm: On an HP EliteBook 6930p laptop, you are mapping the .ISO to drive 0xFF and you see:
QUOTE
DRIVE FF returns 0800h bytes per sector

But later you see the "High Sierra" error?
  Forum: Grub4dos · Post Preview: #103069 · Replies: 44 · Views: 726

happy Birthday, Thunn
Sha0
Posted on: Yesterday, 05:53 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


Yay, birthdays! Yay, Thunn!
  Forum: News · Post Preview: #103065 · Replies: 1 · Views: 43

Boot windows from extended partition
Sha0
Posted on: Yesterday, 05:51 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (Wonko the Sane @ Jun 21 2010, 12:07 PM) *
Right idea thumbsup.gif , maybe wrong example. dubbio.gif
http://support.microsoft.com/kb/249321/en-us

wink.gif

Exactly what do you mean by using the word "wrong"? Is this not exactly the scenario detailed by the article you've referenced?
  Forum: Boot from USB / Boot anywhere · Post Preview: #103064 · Replies: 55 · Views: 2,651

dmio/ftdisk PE1 driver question
Sha0
Posted on: Yesterday, 05:43 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (oriensol @ Jun 21 2010, 11:24 AM) *
@Shao, here're the outputs

In XP

CODE
...
SERVICE_NAME: dmio
        TYPE               : 1   KERNEL_DRIVER
        START_TYPE         : 0   BOOT_START
...

In PE1/Livexp:

CODE
...
SERVICE_NAME: dmio
        TYPE               : 1   KERNEL_DRIVER
        START_TYPE         : 3   BOOT_START
...

Hmmm... What might cause the difference in START_TYPE here? We know that NTLDR loads START_TYPE 0 drivers noted in the Registry. We know that SETUPLDR loads drivers based on TXTSETUP.SIF.
QUOTE (oriensol @ Jun 21 2010, 11:24 AM) *
...Incidentally, the Start Type is 0 in SETUPREG.HIV in both cases (both drivers! not XP and PE1 smile.gif )...

I do not believe that SETUPREG.HIV has any influence over what drivers SETUPLDR loads, but could be mistaken. That is, START_TYPE 0 in SETUPREG.HIV would not help the driver to be loaded at boot time, before the system volume is established. Maybe that's incorrect. If correct though, it explains your 3s as well as a late-start service getting a LEGACY_ entry.

Does that help?
  Forum: Windows 2K/XP/2003 · Post Preview: #103062 · Replies: 25 · Views: 297

Boot windows from extended partition
Sha0
Posted on: Yesterday, 04:08 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


Imagine this: You're Windows. You have kept a note in the Registry for what volume is associated with C:. Suppose your note includes the following volume information:
  • Disk signature
  • Offset to partition
  • Length of partition

Now suppose any of these details change. Now try booting Windows. Windows says, "Hey, my C: volume is missing. Which volume is C:, in that case?" Now suppose Windows looks at the MBR and enumerates the volumes it finds there. Now suppose it picks one to be C:. Suppose it picks the wrong one. Suppose that you log in and userinit.exe is executed, but it's trying to access OS files for the wrong OS installation. Due to the severity of the error, it logs you out and returns you to the logon screen. This is easily reproducible when fiddling with partition shuffling.
  Forum: Boot from USB / Boot anywhere · Post Preview: #103051 · Replies: 55 · Views: 2,651

dmio/ftdisk PE1 driver question
Sha0
Posted on: Yesterday, 03:59 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


oriensol: If you are saying that dmio does not have a LEGACY_ entry in SETUPREG.HIV, but you are seeing one after booting, would you kindly post the output of:
CODE
sc qc dmio

?

Wonko the Sane: Agreed. gdisk is nice. You can have a partition with 0xEE type under XP, certainly. I only mention it for the case where folks are using an XP PE to work with disks that might have only one partition, which is that type, but are expecting to see an NTFS partition somewhere since the computer boots to one.
  Forum: Windows 2K/XP/2003 · Post Preview: #103050 · Replies: 25 · Views: 297

ELTORITO.SYS and the CDROM error
Sha0
Posted on: Yesterday, 03:38 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


I'm lost... Are you getting the "High Sierra" error or not? You showed how QEmu was ok with drive 0xFF, but not with drive 0xA0. Are you saying that the HP EliteBook 6930p is not ok with drive 0xFF, or what? Sorry.
  Forum: Grub4dos · Post Preview: #103048 · Replies: 44 · Views: 726

dmio/ftdisk PE1 driver question
Sha0
Posted on: Yesterday, 03:09 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


oriensol: Leave the entries. For your problem, check that those services are actually started in the PE with sc query dmio, etc. Also beware of GPT-formatted disks, which are becoming more common and are not supported by XP. Zeroing out the MBR sector will destroy all partitioning information but allow for the disk to be re-initialized.
  Forum: Windows 2K/XP/2003 · Post Preview: #103043 · Replies: 25 · Views: 297

Sha0
Posted on: Yesterday, 03:03 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (Lama @ Jun 21 2010, 08:44 AM) *
...doesn't this also mean that there is a way to fool system (or any OS) that boot started from disc instead of partition? without driver? huh.gif...

No. Whatever the strategy is, the OS needs a driver to access wherever the dumped disc's files are. Perhaps the files are extracted to a partition? Perhaps the Mac OS X disc image is dumped as an .ISO, then portions overwritten such as the startup script, which could then find the .ISO, mount it, then continue booting from the mounted image? The OS needs drivers for all of this. If the OS is not an INT 0x13 consumer, the OS needs a driver.
  Forum: FileDisks/RamDisks: Firadisk, etc. · Post Preview: #103042 · Replies: 3 · Views: 160

deploy multiple disk-images w/o user interaction using PXE
Sha0
Posted on: Yesterday, 02:52 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (cgoehl @ Jun 21 2010, 07:28 AM) *
Hi, it seems like I have a very special use case for PXE. At least I could not find any article, which describes whether it has been done yet.
...
Anyone ever done something like that?

There are a few commercial products for this. Symantec Ghost is one of them. Look for "re-imaging products" on an Internet search engine and you should find some. It seems that you are trying to develop your own. Is that right?
QUOTE (cgoehl @ Jun 21 2010, 07:28 AM) *
...Also, I am not sure about the choices for the default case. When there is no image to deploy, the PC should boot from hard disk. Is it better to make PXE timeout or to deploy a pxelinux, which chain-boots the hard drive?

PXELINUX does indeed have MAC address-based config-files as possibilities. See syslinux/doc/pxelinux.txt If you remove a client's MAC address-based config-file, the client will fall back to some default file, which could indeed chain to the local HDD. This approach seems reasonable, as long as the client does not race with the server and reboot before the file is written/removed.
  Forum: LAN and any other methods · Post Preview: #103041 · Replies: 2 · Views: 83

ELTORITO.SYS and the CDROM error
Sha0
Posted on: Yesterday, 02:43 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (Technotika @ Jun 21 2010, 04:31 AM) *
...ELTORITO DRIVER written by...blah blah
DRIVE FF returns 0800h bytes per sector
DEVICE NAME:CD0
Driver installed

DRIVE FF is the confirmation that ElTorito.Sys is using that drive number. If troubleshooting, it's handy to confirm that against whatever your map line was in GRUB4DOS.
QUOTE (Technotika @ Jun 21 2010, 04:31 AM) *
HAVE NOW also been asked to inlcude the hpbranding tools (sets serial names into replaced mainboards etc) iso on the USB stick.

This sounds familiar; a USB stick with all of these utilities on it. Also handy to be able to offer the entire system via a PXE menu system, so you can do all that magic without carrying even a USB stick.
QUOTE (Technotika @ Jun 21 2010, 04:31 AM) *
...IT WORKED!!!!!!?????? what does this mean, if in this scenario the virt CD is accessed ok maybe it isnt hardware, but look at that name \D:mscd001 in the other problem its CD01, is there something in that???look 4 ward to your musings smile.gif

MSCD001 is just a name. You will likely see the same name in AUTOEXEC.BAT. It's just a way for the AUTOEXEC.BAT line to use the same device as was setup using whatever driver in CONFIG.SYS. You could replace both with "foo" or "bar".

Did you pass along the drive number being used by ElTorito.Sys in the non-working scenario, in your post above?
  Forum: Grub4dos · Post Preview: #103039 · Replies: 44 · Views: 726

ISO mapped as C: or barking up wrong tree?
Sha0
Posted on: Yesterday, 02:29 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (Technotika @ Jun 20 2010, 07:43 PM) *
REM LH MSCDEX.EXE /D:cd0 /L:X

What does the /L:X switch do? Have you used the SUBST command before? What drive will get the letter C: if you hide all hard disks?
  Forum: Grub4dos · Post Preview: #102996 · Replies: 4 · Views: 136

FiraDisk (0.0.1.24)
Sha0
Posted on: Yesterday, 02:19 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


Passing information from GRUB4DOS has been discussed here[1] as well as karyonix considering patching GRUB4DOS after some discussion regarding "the right way" happens here[2].

[1] WinVBlock
[2] Feature request: Non-contiguous map and more information in int13 handler
  Forum: FileDisks/RamDisks: Firadisk, etc. · Post Preview: #102995 · Replies: 311 · Views: 71,981

File NON Contiguous problem and solution
Sha0
Posted on: Jun 20 2010, 07:37 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


To clarify, I meant once after each copy operation. As in, you copy to USB storage, you contig it on that USB storage. As long as no-one attempts to defragment their USB storage (uhh...), then the file should remain contiguous on that USB storage.
  Forum: Grub4dos · Post Preview: #102891 · Replies: 14 · Views: 360

Load ISO with MEMDISK using HTTP not TFTP
Sha0
Posted on: Jun 20 2010, 07:34 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


Your VirtualBox VM's NIC might or might not have a gPXE-native driver, which might perform better. If you find its PCI VENdor and DEVice IDs, you could find out. If you hit Control-B once gPXE loads, you should get a gPXE CLI. At the CLI, you can type:
CODE
show net0/busid

to see about the IDs.
  Forum: Hello world! · Post Preview: #102890 · Replies: 17 · Views: 567

Grub4Dos and floppy image size
Sha0
Posted on: Jun 20 2010, 04:41 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


As far as I know, SYS writes the boot code and copies the two files. For different versions of DOS, the boot sector code and these filenames can be different.

I do not believe that file attributes matter; that check would be a waste of space in the < 512-byte space available for boot-code.

Does Murali's script not copy all the contents (including IO.SYS and MSDOS.SYS, or non-MS counterparts) of the source floppy to the destination? It sure looks like it:
CODE
# Copy contents from old image to new image
mount -o loop $NEWIMAGE $MNTNEW
mount -o loop $OLDIMAGE $MNTOLD
cp -r $MNTOLD/* $MNTNEW
umount $MNTNEW
umount $MNTOLD

The only case I know of where the physical location of a file is hard-coded into the boot code is the case of SYSLINUX's LDLINUX.SYS.
  Forum: Grub4dos · Post Preview: #102868 · Replies: 13 · Views: 669

Grub4Dos and floppy image size
Sha0
Posted on: Jun 20 2010, 03:58 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (xpt @ Jun 19 2010, 10:29 PM) *
This requires that the file is continuous, right?

Contiguity is a requirement for GRUB4DOS' sector-mapped disks, but not for RAM disks. GRUB4DOS provides sector-mapped disks (plain map) as well as RAM disks (map --mem). MEMDISK only provides RAM disks.
QUOTE (xpt @ Jun 19 2010, 10:29 PM) *
...
- grub4dos does not need to be provided with extra parameters, but need the file to be continuous to use the (better) int13 approach.

GRUB4DOS and MEMDISK both hook INT 0x13 with any kind of mapping.
QUOTE (xpt @ Jun 19 2010, 10:29 PM) *
- memdisk does not need the file to be continuous but need to provide the (extra) correct parameters.

The floppy parameter is sufficient. The geometry detection should hopefully work for your needs.
QUOTE (xpt @ Jun 19 2010, 10:29 PM) *
...I actually prefer http://people.cs.uchicago.edu/~gmurali/gui/isodos.html
'cause it does not need any manual tweaking.

Thanks for sharing, but...
QUOTE
Creating such a non-standard floppy image is not that difficult. One must ensure that the boot sector of this image is setup correctly. The boot sector created by mkdosfs, will not help. So we need to copy the boot sector from a bootable MSDOS disk (or diskimage). This does not solve the problem completely as well. Parts of the bootsector contain information about the geometry of the floppy. So, we need to copy all the parts of the boot sector which are unrelated to the geometry from the bootable MSDOS disk (or diskimage) and keep the rest of the information untouched.

You can write a batch file/script to do just about anything except your laundry. A batch file could be written for the Windows method given in my previous post, too.
QUOTE (xpt @ Jun 19 2010, 10:29 PM) *
...I want to make a DOS7.10 bootable image, how to make it bootable?

Normally the DOS SYS command is used for this. You might note that Windows XP will format the superfloppy to look for NTLDR, which you can replace with GRLDR (which can then chain IO.SYS). Please note that Murali's script is supposed to take care of copying the relevant boot code from a sample DOS floppy image to your superfloppy image. Doesn't it?
  Forum: Grub4dos · Post Preview: #102865 · Replies: 13 · Views: 669

ELTORITO.SYS and the CDROM error
Sha0
Posted on: Jun 20 2010, 02:43 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


QUOTE (tinybit @ Jun 19 2010, 10:09 PM) *
What I can imagine is, grub4dos will keep the original CDROM drive number(no matter what number it is) of Qemu as it is and only add a new one at (0xA0), whereas memdisk might have shut off the CDROM drive number provided by the virtual machine.

MEMDISK does steal all INT 0x13, AH=0x4B calls (I wrote the El Torito portions). I've been using Bart Lagerweij's ETDump and ETTool to verify.
QUOTE (tinybit @ Jun 19 2010, 10:09 PM) *
Yes, qemu will create a no-emulation-mode cdrom drive even when there was no cdrom configured.

Yes. My QEmu has a native El Torito at 0xE0.
QUOTE (tinybit @ Jun 19 2010, 10:09 PM) *
And then, eltorito.sys will skip the cdrom(provided by the original BIOS) and go successfully.
But in the case of grub4dos, eltorito.sys failed to locate the (0xA0), because the probing might stop at somewhere in the range (0xFF)...(0xA1), when it hit the cdrom drive number provided by the original BIOS.

I'm not 100% sure about this. I've tested successfully drives 0x9F, 0xA0, 0xA1, 0xDF, 0xE0, 0xE1, 0xFE, 0xFF.
QUOTE (tinybit @ Jun 19 2010, 10:09 PM) *
So it appears that this is a bug in eltorito.sys.

Thanks, tinybit. I'm still looking into it.
  Forum: Grub4dos · Post Preview: #102863 · Replies: 44 · Views: 726

Load ISO with MEMDISK using HTTP not TFTP
Sha0
Posted on: Jun 20 2010, 02:31 AM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


It could be a worth-while exercise, in my opinion. You might also consider testing another NIC, just to see if there's a difference. gpxelinux.0 = undionly.kkpxe + pxelinux.0, so it uses the UNDI driver. If your NIC is supported natively by gPXE, it's possible that a gPXE driver for your NIC might perform better. It might be worth-while checking that possibility out, in case the NIC's UNDI from the NIC's vendor is where the performance bottle-next is.
  Forum: Hello world! · Post Preview: #102861 · Replies: 17 · Views: 567

File NON Contiguous problem and solution
Sha0
Posted on: Jun 19 2010, 11:22 PM


Frequent Member
***

Group: Advanced user
Posts: 404
Joined: 1-August 07
From: Boot-Land Forums
Member No.: 9,903


Good luck finding a boot-loader without this requirement. As long as the file's been contig'd once, there's not really a good reason for it to become fragmented thereafter. Contiguity is not a requirement for RAM disks, but you don't have 8+ GB of RAM, I'd bet, nor the time to load it.
  Forum: Grub4dos · Post Preview: #102856 · Replies: 14 · Views: 360

16 Pages V   1 2 3 > » 

New Posts  New Replies
No New Posts  No New Replies
Hot topic  Hot Topic (New)
No new  Hot Topic (No New)
Poll  Poll (New)
No new votes  Poll (No New)
Closed  Locked Topic
Moved  Moved Topic