Ubuntu8.04 XEN3.4.2 installation process FAQs

Test Environment Description: I use VMware virtual installation Ubuntu8.04 on an XP host and use xen3.4.2 as DOM0. The following problems may occur: 1 common problems and solutions 1.1Python version after modifying the configuration prompt python error, possible cause: when installing the xen-3.4.2, require the use of python2.6, but the total error during compilation, then changed to python2.5; after the system runs normally, you need to change python to python2.6 to support other application tools such

Test Environment Description:
I use VMware to install Ubuntu 8.04 on an XP host and use xen3.4.2 as DOM0. The following problems may occur:
1. FAQs and Solutions
1.1 Python version
A python error is prompted after the configuration is modified. Possible cause: python2.6 is required when the xen-3.4.2 is installed, but the total error occurs during compilation, which is later changed to python2.5; after the system is running properly, you need to change python to python2.6 to support other application tools, such as virt-manager of a later version. After the system is installed, versions 2.5 and 2.6 should be automatically supported. When xend is used, you can create a link to start version 2.5 by default. You can use this command to switch versions:
# Rm/usr/bin/python
# Ln-s/usr/bin/python2.6/usr/bin/python
# Python
Press enter to find that the version is 2.6

1.2 Xend startup Error
1.2.1 prompt brctl
Bash: brctl: command not found
To install the required package online, run apt-get install bridge-utils to install the bridge-utils package;
1.2.2 Ifdown prompt
When the Client starts xend, it cannot bridge eth0 (The Xend service is normal, but there is no bridge). The prompt is as follows:
$ Xend restart
Ifdown: interface eth0 not configured
SIOCSIFNAME: Device or resource busy
Possible cause: the logic of DRBL is to first start the network device (started in the initrd phase) to mount NFS as rootfs, but ifdown eth0 will first be triggered when Xend is started, start eth0 after performing some brctl-related actions. Currently, the following solutions are available:
1): In addition to the default information, eth0 is added to/etc/network/interface.
Auto lo
Iface lo inet loopback
Auto eth0
Iface eth0 inet static
Address 192.168.255.129
Netmask 255.255.255.0
Network 192.168.20.0
Gateway 192.168.20.2
Then you need to restart the network service to start xend:
Root @ ubuntu8 :~ # Xend start
Ifdown: interface eth0 not configured
SIOCSIFNAME: Device or resource busy
Root @ ubuntu8 :~ # Service networking stop
* Deconfiguring network interfaces... [OK]
Root @ ubuntu8 :~ # Service networking start
* Refreshing ing network interfaces... [OK]
Root @ ubuntu8 :~ # Xend start
Waiting for peth0 to negotiate link.

2): manually ifup eth0 first, causing the entire system to stop (because the network is stopped, rootfs cannot be mounted, and the terminal cannot be controlled)
3): Use the alias $ xend restart instead.
Error: either "local" is duplicate, or "eth0" is a garbage.
4): Modify/etc/xen/script/network-bridge.
Instead of using the default bridge, modify the network configuration and Script: network-bridge to network-NAT in/etc/xen/xend-conf.cfg; change vif-bridge to vif-nat.
However, the following prompt does not affect xend usage:
+ Iptables-t nat-a postrouting-o eth0-j MASQUERADE
FATAL: Module ip_tables not found.
Iptables v1.3.8: can't initialize iptables table 'nat ': iptables who? (Do you need to insmod ?)
Perhaps iptables or your kernel needs to be upgraded.
5): We recommend that you use netwrok-route instead if you use NFS roots, just to obtain the dhcp ip address.
Based on our experience, we currently use methods 1 or 4.
Note:
After the system starts, the default network interface number changes, for example, from eth0 to eth1. The same is true in DomainU. You need to modify eth1 in interfaces in/etc/network and then start the network service according to the change:
# Service networking start
Modify the configurations in interfaces back to eth0. After the machine is restarted, the network service is normal.
The reason may be related to the operation on the network interface when xend is started.
1.3 VNC Interface
1.3.1 gray and not operable
Problem: the gray window appears when VNCSERVER/VNCVIEWER is used to connect to dom0.
Solution: under root, modify the./vnc/xstartup file and replace the last line of twm with the following two lines:
Unset SESSION_MANAGER
Gnome-session &
1.3.2 gray but operable
When you connect to the VNCSERVER, the page displays gray or prompts which channel has an error, you need to re-open the VNCSERVER:
# Vncserver-kill: x
Use this command to close the related desktop and then open it.
1.3.3 connection prompt channel error
When you connect to the VNCSERVER, the page displays gray or prompts which channel has an error, you need to re-open the VNCSERVER:
# Vncserver-kill: x
Use this command to close the related desktop and then open it.
1.4 VNC shared by multiple users
How can the host desktop in DomainU be shared by multiple users at the same time? Currently, only one access terminal can be set using vncdisplay = n in xxx. hvm under/etc/xen.
Solution:
1. For windows DOMU, you can use remote desktop or install vncserver on the domu system, and configure to allow access from an IP segment or some IP addresses.
2. If it is a linux domu, you need to install VNCSERVER and configure it to enable the desktop at startup.
1.5 Virt-manager
1.5.1 Installation Problems
Under normal circumstances, you can directly use apt-get install virt-manager to install virt. If the library is missing, you should continue to add the installation-such as libvirt;
1. Start xend first;
2. connect to connect xen in virt-manager;
3. If an error occurs, the possible causes are as follows:
1) xend is not started; or xend is incorrectly configured.
In the/etc/xen/xend-config.sxp file, enable (xen-api-server (unix), (xend-http-server yes) /(xend-unix-server yes) --- enable one of the two servers;
2) libvirt library is not installed successfully-virt-manager is mainly used to monitor and manage xend Based on libvirt.
4. debug the libvirt Library:
In the test program, the-lvirt option should be used during compilation to link to the libvirt-dev library. If the undefined reference to xxx error message appears, it may be because the library libvirt-dev is not installed, use apt-get install libvirt-dev for installation.
1.5.2 startup Problems
Virt-manager startup problems:
Unable to open a connection to the Xen hypervisor/daemon.
Verify that:
-A Xen host kernel was booted
-The Xen service has been started
Details:
Unable to open connection to hypervisor URI 'xen :///':
VirConnectOpenReadOnly () failed cocould not use Xen hypervisor entry/tmp/liyun_proxy_conn
Traceback (most recent call last ):
File "/usr/share/virt-manager/connection. py", line 332, in _ open_thread
Self. vmm = libvirt. openReadOnly (self. uri)
File "/usr/lib/python2.5/site-packages/libvirt. py", line 144, in openReadOnly
If ret is None: raise libvirtError ('virconnectopenreadonly () failed ')
LibvirtError: virConnectOpenReadOnly () failed cocould not use Xen hypervisor entry/tmp/liyun_proxy_conn
Solution:
Same problem here with the same kernel amd64. I tried enabling these two settings in xend-config.sxp, but without auccess :( configure in xend-config :)
(Xen-api-server (unix )))
(Xend-unix-server yes)
1.5.3 online installation toolkit Problems
Tip:
Root @ ubuntu8 :~ # Apt-get install bridge-utils
E: cocould not get lock/var/lib/dpkg/lock-open (11 Resource temporarily unavailable)
E: Unable to lock the administration directory (/var/lib/dpkg/), is another process using it?
Solution:
Because another download and installation process does not work at the same time, you need to disable or wait until another download and installation process ends, or restart the system.
Personal test, for reference only!