Oluf Lorenzen wrote:
>> This is interesting.  I run a few dozens of virtual machines here,
>> all with virtio disk, and experiment alot with it too, but I never
>> had this problem, and don't know anyone had it in #kvm channel
>> either.  Please provide complete kvm command line.
> 
> Here you are:
> /usr/bin/qemu-kvm -S -M pc-0.12 -enable-kvm -m 100 \

I'll try to work from this on, but I've 2 immediate questions.
What is /usr/bin/qemu-kvm ?  Debian does not ship such binary,
neither in kvm nor in qemu-kvm nor in qemu packages.  The
binary we ship is /usr/bin/kvm not /usr/bin/qemu-kvm.

>     -smp 1,sockets=1,cores=1,threads=1 \
>     -name squeeze-test \
>     -uuid bfc456ea-f3bc-d4f3-9d02-dd0511751f7c \
>     -nodefaults \
>     -chardev 
> socket,id=monitor,path=/var/lib/libvirt/qemu/squeeze-test.monitor,server,nowait
>  \
>     -mon chardev=monitor,mode=readline \
>     -rtc base=utc -boot c \
>     -drive 
> file=/var/lib/libvirt/images/squeeze-test.img,if=none,id=drive-virtio-disk0,format=raw
>  \
>     -device 
> virtio-blk-pci,bus=pci.0,addr=0x5,drive=drive-virtio-disk0,id=virtio-disk0 \

So, this is using virtio and -boot c.  qemu-kvm-0.12 isn't able
to boot from virtio disk this way, it needs ",boot=on" device
property instead of "-boot c".  The ability to treat virtio
drives as ide ones when booting is concerned (i.e. in bios)
first appeared in 0.13 (counting only official releases).

So it appears libvirt calls it incorrectly.  Libvirt is
too new?  What are you using there - squeeze? sid?
What libvirt version?

>     -device 
> virtio-net-pci,vlan=0,id=net0,mac=52:54:00:e4:19:c3,bus=pci.0,addr=0x3 \
>     -net tap,fd=43,vlan=0,name=hostnet0 -chardev pty,id=serial0 \
>     -device isa-serial,chardev=serial0 -usb \
>     -vnc 127.0.0.1:4 -vga cirrus \
>     -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x6
> 
> The configfile is at <http://paste.grml.org/120/>.
> 
>> What tells you it is not added to debian?
> 
> Well, i had a look at the files from the qemu-kvm-source-package
> looking for optionrom/signrom.sh, and the lines 41/42 did not contain
> the line from the patch.

Yes it does not contain that line, but the surrounding
lines are changed too.

> I am not a programmer so i might be completely on the wrong path about this ;)

Ok, I see, no problem.

Thanks!

/mjt



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Reply via email to