Windows XP SP3 not booting up after moving VDI image
Posted: 27. Nov 2010, 00:55
I had two properly functioning Windows XP SP3 VM's. Then I moved the VDI's to a new partition and they wouldn't work. But, they do now. Here's everything that happened to the best of my knowledge. I add this disclaimer because I've been working on this problem on and off for about six hours (and several cups of coffee) now and may have possibly left out some things. As well as encountering a number of other unrelated problems with my host OS Linux installation during this time. So, here goes.
When I booted up I got a 0x7B BSOD STOP message. So, Various sources told me that it was an IDE controller device. So, after some searching Microsoft's website told me to do an in place upgrade of the OS. Which will supposedly reinstall the OS but leave my files on it alone. Before attempting the repair install (which, I never actually ended up using. I just went to the recovery console) I tried everything I could think of, even removing the IDE controller altogether, to no avail.
But, this is where the real kicker starts. The problem seemed to run deeper than just an IDE controller issue. The Windows XP installation CD couldn't find a hard drive installed on the computer. Although, an Ubuntu live CD I booted off of could. A possible solution I found (I don't have the exact link anymore) on this forum involved downloading some Intel HD drivers, putting them on a floppy disk, and loading them up during the XP installation process.
I loaded up one of the Intel drivers, went through the process, typed in my admin password, got to the recovery console, and typed in "fixmbr /r". Then, when I got to boot up the OS I got "FATAL: can not red from the boot medium! system halted". After this happened, I brought up the boot loader (as I had done in the past) and now the SATA drives were showing up in the list of bootable mediums, which they had not done before. But, I still couldn't boot the OS off of them.
After this, I decided to simply try removing the VDI from the SATA/ACHI controller and creating a new IDE controller from which to mount the VDI to. Which worked on both VM's. Too bad I hadn't came up with this idea hours ago. Now I have two questions:
1) In real life SATA is faster than IDE. Will running a VM HD from a virtual IDE controller likewise also be slower than running one from a virtual SATA controller? In other words, are virtual SATA and IDE controllers limited to the speeds of their physical counterparts?
2) What the heck just happened? Why would XP not boot from a virtual SATA controller when that's what it was installed the the VDI using. But, it would instantly boot from an IDE even though it's never accessed the VDI from a virtual IDE controller.
When I booted up I got a 0x7B BSOD STOP message. So, Various sources told me that it was an IDE controller device. So, after some searching Microsoft's website told me to do an in place upgrade of the OS. Which will supposedly reinstall the OS but leave my files on it alone. Before attempting the repair install (which, I never actually ended up using. I just went to the recovery console) I tried everything I could think of, even removing the IDE controller altogether, to no avail.
But, this is where the real kicker starts. The problem seemed to run deeper than just an IDE controller issue. The Windows XP installation CD couldn't find a hard drive installed on the computer. Although, an Ubuntu live CD I booted off of could. A possible solution I found (I don't have the exact link anymore) on this forum involved downloading some Intel HD drivers, putting them on a floppy disk, and loading them up during the XP installation process.
I loaded up one of the Intel drivers, went through the process, typed in my admin password, got to the recovery console, and typed in "fixmbr /r". Then, when I got to boot up the OS I got "FATAL: can not red from the boot medium! system halted". After this happened, I brought up the boot loader (as I had done in the past) and now the SATA drives were showing up in the list of bootable mediums, which they had not done before. But, I still couldn't boot the OS off of them.
After this, I decided to simply try removing the VDI from the SATA/ACHI controller and creating a new IDE controller from which to mount the VDI to. Which worked on both VM's. Too bad I hadn't came up with this idea hours ago. Now I have two questions:
1) In real life SATA is faster than IDE. Will running a VM HD from a virtual IDE controller likewise also be slower than running one from a virtual SATA controller? In other words, are virtual SATA and IDE controllers limited to the speeds of their physical counterparts?
2) What the heck just happened? Why would XP not boot from a virtual SATA controller when that's what it was installed the the VDI using. But, it would instantly boot from an IDE even though it's never accessed the VDI from a virtual IDE controller.