...and the mac address is found from running ifconfig ??? - what MAC address are you using in the vm config file?
Code:
name = 'vm01'
disk = ['phy:/dev/cis-du01/vm01,xvda,w']
vif = ['mac=78:2b:cb:43:39:8e,bridge=xenbr0']
...
You cannot use an existing (= real hardware) MAC address in your vif = configuration !!! You are using the MAC address of your eth0 interface - this is wrong! Your guest connects to the virtual bridge (xenbr0), in which case you have to specify a virtual MAC address (actually, you don't need to specify a MAC address at all, Xen will select a random one for you). Edit your vm01 config file and enter the following vif statement (pay attention to the MAC address):
Code:
vif = [ 'mac=00:16:3e:68:01:01,bridge=xenbr0' ]
You can change the 01:01 part to suit your needs. If you have several VMs running on your machine/network, you should specify a MAC address and make sure to change the 01:01 part to 01:02, 01:03, etc., else you will be in trouble. In fact, it may be good practice to use 01:01, 02:01 for different virtual bridges (or different PCs), and 01:01, 01:02, etc. for different VMs/virtual ports on the same virtual bridge. The 00:16:3e:68 MAC address part has been reserved by Xen for the purpose of creating virtual MAC addresses.
Next I would edit the /etc/network/interfaces file:
1. Comment out auto eth0 and iface eth0... (see below).
2. Add
Code:
dns-nameservers 8.8.8.8
to the auto xenbr0 section (see below).
3. Change netmask 255.255.255.192 to netmask 255.255.255.0 (unless you know that .192 will work!)
4. Unless you are totally sure about the network xxx.32.92.0 line, comment it out.
Save and restart networking, then give it a try.
Here the modified /etc/network/interfaces file:
Code:
# The loopback network interface
auto lo
iface lo inet loopback
# The primary network interface
#auto eth0
#iface eth0 inet manual
auto xenbr0
iface xenbr0 inet static
address xxx.32.92.xx
netmask 255.255.255.0
# network xxx.32.92.0
broadcast xxx.32.92.xx
gateway xxx.32.92.x
dns-nameservers 8.8.8.8
bridge_ports eth0
bridge_fd 9
bridge_hello 2
bridge_maxage 12
bridge_stp off
Explanations:
1. You have a bridged network, so dom0 and every domU connect to the bridge. You don't configure eth0 as the primary network interface, as we use the xenbr0 bridge for the networking. The bridge will use eth0 as the network port (see bridge_ports eth0).
2. In case your PC doesn't resolve names - dns-nameservers 8.8.8.8 should solve that.
3. I would have to look up using network masks, but if any of your masks are incorrect, it won't work. My suggestion is to stick to the most common options, such as netmask 255.255.255.0 and without the network xxx.32.92.0 option, and play with them once you got everything working.
It may not be related to your error, but I would also edit the /etc/xen/xend-config.sxp file and get the settings somewhat in line with this
http://forums.linuxmint.com/viewtopi...112013#p628911. If you want to use virt-manager, you need xend-unix-server yes. The Xen documentation discourages the use of the xend network scripts (see the link for more on network configuration), I believe (network-script /bin/true) is the option.
Here is my xend-config.sxp file for reference:
Code:
#(xend-http-server no)
(xend-unix-server yes)
(network-script /bin/true)
(vif-script vif-bridge)
(dom0-min-mem 1024)
(enable-dom0-ballooning no)
(total_available_memory 0) # You may need to change that!
(dom0-cpus 0)
(vncpasswd '')
#(keymap 'en-us') # You may want to specify this
Yeah, Xen can be a little challenging at the beginning. Hope this solves the issue. Good luck!
Adv Reply
Originally Posted by tgalati4
Bookmarks