When you attempt to boot a Xen DomU on Debian Lenny Dom0, the console might hang as the follows:
Starting NFS servers: nfsd mountd.
Starting automounter: amd.
Starting anac(h)ronistic cron: anacron.
Starting periodic command scheduler: crond.
Running local boot scripts (/etc/rc.local).
This problem occurs when using xen-tools to create a new (Debian Lenny in my case) guest image and configuration.
The cause for the problem is when the guest configuration file does not have the “extra” configured.
To solve this problem – just add -
extra = ‘console=hvc0 xencons=tty‘
To your guest configuration file.
kernel = '/boot/vmlinuz-2.6.18.8-xen'extra = ‘
ramdisk = '/boot/initrd.img-2.6.18.8-xen'
console=hvc0 xencons=tty‘
memory = '512'
root = '/dev/sda2 ro'
disk = [
'phy:/dev/VM/vm01-disk,sda2,w',
'phy:/dev/VM/vm01-swap,sda1,w',
]
name = 'vm01'
vif = [ 'mac=00:16:7E:DC:C9:F6' ]
on_poweroff = 'destroy'
on_reboot = 'restart'
on_crash = 'restart'
If you are using pygrub, edit /boot/grub/menu.lst on the domU and append - console=hvc0 xencons=tty to the kernel line.
I have found that this problem does NOT exists when running Etch and using xne-tools to create an image. I confirm that this happens on my Lenny machine running an amd64 kernel when using xen-tools to create an image.
[...] this post I described the workaround for the hanging at the DomU console when running Dom0 on Lenny. [...]
Thank you thank you thank you!!!!
I have previously also added this line to dom0′s /etc/inittab:
co:2345:respawn:/sbin/getty 38400 hvc0
I’m not sure I need it, but it works so I won’t mess with it anymore.
But your advice solved the problem for me! Thanks again!
thanks for the tip
—
extra = ‘console=hvc0 xencons=tty‘
—