These are some notes I wrote sometime ago.
Installation:
The following packages may need to be installed:
yum update -y
yum install libvirt qemu-kvm virt-manager virt-install
yum install libguestfs-tools
yum install qemu-img -y
From the RHEL6: Virtualization Host Configuration and Guest Installation Guide
yum install qemu-kvm qemu-img
yum install virt-manager libvirt libvirt-python python-virtinst libvirt-client
List all virtual machines:
[ ~]# virsh list --all
Id Name State
----------------------------------------------------
- openstack-compute shut off
- openstack-controller shut off
- openstack-nethost shut off
- openstack-TEMPLATE shut off
- rhel6x-cloud shut off
List network configuration:
[ ~]# virsh net-list
Name State Autostart Persistent
----------------------------------------------------------
default active yes yes
Define a network:
If no "default" (NAT) network is configured:
virsh net-define /usr/share/libvirt/networks/default.xml
List defined virtual networks:
# virsh net-list
Name State Autostart Persistent
----------------------------------------------------------
default active yes yes
Show specific network information:
[ ~]# virsh net-info default
Name default
UUID 99ff5b5f-a04e-498e-ab06-dc1cdd8d544c
Active: yes
Persistent: yes
Autostart: yes
Bridge: virbr0
Edit a network:
To edit network configuration (using default as an example):
virsh net-edit default
Network “default” standard definition looks like this:
<network>
<name>default</name>
<bridge name="virbr0" />
<forward/>
<ip address="192.168.122.1" netmask="255.255.255.0">
<dhcp>
<range start="192.168.122.2" end="192.168.122.254" />
</dhcp>
</ip>
</network>
The Network can be modified by changing IP scheme and many other things.
One trick is to assign IP address based on MAC, and also adjust the pool size or range for dhcp leases:
<network>
<name>default</name>
<uuid>99ff5b5f-a04e-498e-ab06-dc1cdd8d544c</uuid>
<forward mode='nat'/>
<bridge name='virbr0' stp='on' delay='0' />
<mac address='52:54:00:40:bd:27'/>
<ip address='192.168.122.1' netmask='255.255.255.0'>
<dhcp>
<range start='192.168.122.2' end='192.168.122.9' />
<host mac='52:54:00:00:00:01' name='openstack-controller' ip='192.168.122.101' />
<host mac='52:54:00:00:00:02' name='openstack-nethost' ip='192.168.122.102' />
<host mac='52:54:00:00:00:03' name='openstack-compute' ip='192.168.122.103' />
<host mac='52:54:00:00:00:04' name='openstack-controller2' ip='192.168.122.104' />
</dhcp>
</ip>
</network>
Start a network if it’s not running:
virsh net-start default
Stop a network:
virsh net-destroy default
Set a network to auto start:
virsh net-autostart default
Edit a working network definition:
Edit a current running/working network without need to hard restart
[ ~]# virsh net-update default add-last ip-dhcp-host \
> --xml "<host mac='52:54:00:00:00:12' name='server2' ip='192.168.122.102' />" \
> --live --config
Updated network default persistent config and live state
Create a Virtual Image:
Create a virtual image that will boot using an installation image
Example A:
virt-install --name=openstack-controller \
--ram=2000 \
--file=/var/lib/libvirt/images/openstack-controller.img \
--cdrom=/var/lib/libvirt/images/rhel-server-6.5-x86_64-dvd.iso \
--noautoconsole \
--vnc \
--file-size 30 \
--os-type linux \
--os-variant rhel6 \
--network network:default,mac=52:54:00:00:00:01
`
Example B:
```bash
virt-install --name rhel6x-cloud \
--ram 1000 \
--cdrom /var/lib/libvirt/images/rhel-server-6.5-x86_64-dvd.iso \
--noautoconsole \
--vnc \
--os-variant rhel6 \
--network network:default,mac=52:55:00:00:00:01 \
--disk path=/var/lib/libvirt/images/rhel6x-cloud.qcow2,format=qcow2,size=10
Example C:
qemu-img create -f qcow2 -o size=20G,preallocation=metadata rhel6-disk.qcow2
virt-install --name rhel6-guest \
--ram=2048 \
--vcpus=2 \
--cpu host \
--hvm \
--cdrom /var/lib/libvirt/isos/rhel-server-6.6-x86_64-dvd.iso \
--noautoconsole \
--graphics vnc \
--os-variant rhel6 \
--description "RHEL6 KVM guest 1" \
--network networkefault \
--disk path=/var/lib/libvirt/images/rhel6-disk.qcow2,format=qcow2,bus=virtio
Example D:
virt-install --name rhnsat56 \
--vcpus 2 \
--ram 4096 \
--cdrom /var/lib/libvirt/images/rhel-server-6.5-x86_64-dvd.iso \
--noautoconsole \
--os-variant rhel6 \
--network network:default,mac=52:54:00:00:00:04 \
--disk path=/home/VirtualImages/rhnsat56-root.qcow2,format=qcow2,size=20 \
--disk path=/home/VirtualImages/rhnsat56-satellite.qcow2,format=qcow2,size=80 \
--disk path=/home/VirtualImages/rhnsat56-cache.qcow2,format=qcow2,size=10 \
--disk path=/home/VirtualImages/rhnsat56-pgsql.qcow2,format=qcow2,size=20
Example E:
virt-install --name lxc-host-1 \
--vcpus 2 \
--ram 2048 \
--cdrom /var/lib/libvirt/images/rhel-server-6.4-x86_64-dvd.iso \
--noautoconsole \
--os-variant rhel6 \
--network network:default,mac=52:54:00:00:00:04 \
--disk path=/home/VirtualMachines/lxc-host-1.qcow2,format=qcow2,size=20
Prepare VM for cloning:
After the installation is finish and the VM is powered off, if cloning will be needed then it is recommended that the virtual machine is prepared for cloning. For this use virt-sysprep, which resets or unconfigures a virtual machine so that clones can be made from it.
Virt-sysprep is also useful when images will be used on PaaS systems like openstack.
virt-sysprep -d openstack-controller
Another common thing to do for preparing a VM is to configure the ifcfg-eth{x}file to “ONBOOT=yes” so the networking auto starts upon first boot.
virt-edit -d rhel64-cloud /etc/sysconfig/network-scripts/ifcfg-eth0 -e 's/^ONBOOT=.*/ONBOOT="yes"/'
