[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] RE: [Xen-users] Creating a full virtualized Debian domU
Try changing your disk setup to [1]: disk = [ 'phy:/dev/vm/debian.swap,ioemu:sda,w', 'phy:/dev/vm/debian.raiz,ioemu:sdb,w', 'phy:/dev/hda,ioemu:hda:cdrom,r' ] There's a difference between the way that HVM and PV diomains work in disk access. The entire disk is the smallest granularity that the guest can see in HVM, whilst PV allows individual partitions to be exported. To get the same effect as I thinnk you originally intended, you'll need to make one device available to the guest like this: disk = [ 'phy:/dev/vm/debian,ioemu:sda,w', 'phy:/dev/hda,ioemu:hda:cdrom,r' ] Then you create the swap and other partitions inside the guest (technically, you could create them outside, but that's more work to no advantage). It's important to understand that partitions isn't a hardware feature. It's entirely a software functionality. The HVM interface is a hardware-based interface, so only things that can be seen at the hardware level can be differentiated to the guest. Since the PV model is more intrusive inside the kernel (it's got code-modifications inside the kernel and drivers), it's perfectly possible to modify things "further up" in the software chain here. Finally a minor suggestion: Instead of 'boot="dc"', try 'boot="cd"', as that will attempt to boot the hard-disk first, which helps when you're rebooting the domain after the OS has been installed, you don't have to go change the configuration before it boots the actual installed OS. [1] I'm not entirely sure if SCSI-type devices are allowed either. If the proposed change still doesn't work, try replacing sd with hd and see if that works better. Please let me know if "sd" devices work, as I'd like to know. -- Mats > -----Original Message----- > From: xen-users-bounces@xxxxxxxxxxxxxxxxxxx > [mailto:xen-users-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf Of > Daniel Bareiro > Sent: 24 April 2007 18:29 > To: Xen-Users > Subject: [Xen-users] Creating a full virtualized Debian domU > > I'm getting some problems to create a HVM Debian domU. > > HVM in enabled: > > (XEN) AMD SVM Extension is enabled for cpu 0. > > This is the domU's config file: > > xenhost6:/etc/xen# cat debian.hvm|grep ^[^#] > kernel = "hvmloader" > builder='hvm' > memory = 256 > shadow_memory = 8 > name = "Debian" > vif = [ 'type=ioemu, bridge=xenbr1' ] > disk = [ 'phy:/dev/vm/debian.swap,ioemu:sda1,w', > 'phy:/dev/vm/debian.raiz,ioemu:sda2,w', > 'phy:/dev/hda,ioemu:hda:cdrom,r' ] > device_model = 'qemu-dm' > boot="dc" > sdl=1 > vnc=1 > stdvga=0 > serial='pty' > > > xenhost6:/etc/xen# xm create -c debian.hvm > Using config file "debian.hvm". > Started domain Debian > > It freeze here. > > In another console: > > the output from xm dmesg is: > > (XEN) (GUEST: 5) HVM Loader > (XEN) (GUEST: 5) Detected Xen v3.0.3-1 > (XEN) (GUEST: 5) Writing SMBIOS tables ... > (XEN) (GUEST: 5) Loading ROMBIOS ... > (XEN) (GUEST: 5) Loading Cirrus VGABIOS ... > (XEN) (GUEST: 5) SVM go ... > (XEN) (GUEST: 5) rombios.c,v 1.138 2005/05/07 15:55:26 vruppert Exp $ > (XEN) (GUEST: 5) VGABios $Id: vgabios.c,v 1.61 2005/05/24 16:50:50 > vruppert Exp $ > (XEN) (GUEST: 5) HVMAssist BIOS, 1 cpu, $Revision: 1.138 $ $Date: > 2005/05/07 15:55:26 $ > (XEN) (GUEST: 5) > (XEN) (GUEST: 5) ata0 master: QEMU CD-ROM ATAPI-4 CD-Rom/DVD-Rom > (XEN) (GUEST: 5) ata0 slave: Unknown device > (XEN) (GUEST: 5) > (XEN) (GUEST: 5) Booting from CD-Rom... > > and the domU appears to be up: > > xenhost6:~# xm list > Name ID Mem(MiB) VCPUs > State Time(s) > Debian 5 256 1 > ------ 734.1 > Domain-0 0 1712 1 > r----- 33.7 > > but the Debian installer doesn't start. Any idea? > > Thanks in advance. > > Regards, > Daniel > -- > Daniel Bareiro - System Administrator > Fingerprint: BFB3 08D6 B4D1 31B2 72B9 29CE 6696 BF1B 14E6 1D37 > Powered by Debian GNU/Linux Etch - Linux user #188.598 > _______________________________________________ Xen-users mailing list Xen-users@xxxxxxxxxxxxxxxxxxx http://lists.xensource.com/xen-users
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |