01.05.2010 10:05, Martin Stut wrote:
The system run attempt kvm -m 1024 -hda /backup/qemu/francke4/francke4.img -cdrom /backup/qemu/francke4/winnt4.iso -cpu host -boot order=cd -net nic,model=rtl8139,vlan=0,macaddr=00:11:22:33:44:66 -net user,vlan=0 -localtime -name francke4 -monitor stdio results in => BSOD stop 0x0000001e in ntoskrnl.exe
..which is, again, "inaccessible boot device". I tried installing winNT SP1 in kvm-72 and upgrading kvm to 0.12 as available in bpo and testing. The guest system boots and works just fine here. So I can only conclude that the change in question was in earlier kvm, i.e., when your guest were installed it saw different hardware, which now does not work anymore. If this is the case, there's very little reason to dig into this further, I'm afraid. Alternative possible cause is that you're using even older NT, for which the problem will be reproduceable when upgrading from kvm-72 to kvm-0.12 (note again I'm using SP1 as base install). But after a bit of thinking, it looks like fixing this isn't worth the effort actually: it is how winNT reacts to changes in the hardware it is running on, give or take bugs fixed in later service packs. And by changing kvm you _are_ changing the virtual "hardware". I just looked at the winNT hardware configuration - it is already using some sort of "generic IDE" driver for the IDE controller for me (it is listed in the "SCSI controllers" applet in the Control Panel). So I don't know what else to try, actually. In order to move win2000+ one usually switches ide to a "generic IDE" driver, so it works on any IDE controller. With winNT we've only one driver, which already is generic. Oh well. /mjt -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
