Installing the CSR 1000V on a KVM Hypervisor
The process for installing the CSR 1000V on a KVM hypervisor has two phases:
Bring up the VM with the CSR 1000V on ESXi.
Connect the vNIC with the CSR 1000V.
Bring Up the CSR 1000V as a Guest
Follow these steps to update essential packages on a Linux managed server so it can work as a type 1 hypervisor and run a CSR 1000V VM:
Step 1. Install the VM packages virt-manager, qemu-kvm, and bridge-utils like this:
apt-get install virt-manager apt-get install qemu-kvm
apt-get install bridge-utils
or like this:
yum install virt-manager yum install qemu-kvm
yum install bridge-utils
Figure 4-30 shows the installation of packages required for CSR creation.
Figure 4-30 Package Installation on a KVM Hypervisor
Step 2. Launch Virtual Machine Manager, which is the front end to KVM/QEMU that allows installation and management of CSR VMs, by selecting Application, System, Virtual Machine Manager.
Click the Create a New Virtual Machine icon, and the dialog shown in Figure 4-31 appears. Click the Forward button.
Figure 4-31 Creating a Guest VM
Step 3. Load the ISO image (which you download from software.cisco.com ) for the CSR 1000V, as shown in Figure 4-32. Click the Forward button.
Figure 4-32 ISO Image Bootup for the CSR 1000V
Step 4. Allocate hardware resources for the guest VM as shown in Figure 4-33. (Refer to Table 2-2 in Chapter 2 for further allocation information.) Click Forward.
Figure 4-33 Choosing Memory and CPU Settings
Step 5. Select hardware resources, as shown in Figure 4-34, and click Forward.
Figure 4-34 Selecting Hardware Resources
Step 6. Look over the hardware resources summary (see Figure 4-35) and make any changes needed. Click Finish.
Figure 4-35 Resources Summary Snapshot
Step 7. To apply changes for the guest VM, select Application, System, Virtual Machine Manager and highlight the CSR installed in the VMM. Then click the Show Virtual Hardware Details tab and click the Add Hardware button, as shown in Figure 4-36.
Figure 4-36 Applying Hardware VM Changes
Step 8. To create serial connection access for console access, select Serial, and then select TCP for Device Type and provide the telnet information, as shown in Figure 4-37.
Figure 4-37 Creating the Serial Interface
Step 9. In the Virtual Machine Manager, highlight the guest VM and shut it down (if it is not down already). (See Figure 4-38.)
Figure 4-38 Shutting Down the Guest VM
The guest VM goes down, as shown in Figure 4-39.
Figure 4-39 Shutdown of the Guest VM
Step 10. Access the router from the console, as shown in Figure 4-40. Make sure the VM is powered up before you try to access it.
Figure 4-40 Console Access to the KVM
Step 11. Use the serial interface command for telnet access: platform console serial and write mem, as shown in Figure 4-41.
Figure 4-41 Router Console for Telnet Access
Step 12. Access the CSR 1000V via the telnet, as shown in Figure 4-42.
Figure 4-42 Telnet Connection to the CSR 1000V
Step 13. Ensure that your virtual machine is shut down, and then start vNIC provisioning by selecting Show Virtual Hardware Details, NIC, as shown in Figure 4-43.
Figure 4-43 Accessing CSR 1000V Network Settings
Step 14. In the Virtual Machine Manager, select virtio as the device model (see Figure 4-44) because it is the para-virtualized driver in Linux. Using virtio is the best way to exploit the underlying kernel for I/O virtualization. It provides an efficient abstraction for hypervisors and a common set of I/O drivers.
Figure 4-44 Selecting CSR 1000V Network Settings
Select the virtual network with NAT to tie all VMs in the same bridge domain and NAT it to the outgoing physical interface (see Figure 4-45). Attach the other NIC to the bridge tap.
Figure 4-45 CSR 1000V NIC Settings
In KVM, macvtap is a combination of the macvlan driver and a Tap device. Here the function of the macvlan driver is to create virtual interfaces and map virtual interfaces to physical network interfaces. A unique MAC address identifies each virtual interface to the physical interface. A TAP interface is a software only interface that exists only in the kernel. You use Tap interfaces to enable user-space networking and allow passing of datagrams directly between VMs instead of sending datagrams to and from a physical interface. The macvtap interface combines these two functions together (see Figure 4-46).
Figure 4-46 macvtap Diagram
Step 15. Configure the mapping of the vNIC to the physical interface:
Access the directory /etc/network/interfaces/ifcfg-br0 on the Ubuntu host and view the bridge type (see Figure 4-47).
Figure 4-47 Bridge Configuration File Output
Access the directory /etc/network/interfaces/ifcfg-eth4 and configure the vNIC to be in the same bridge type, BR0 (see Figure 4-48).
Figure 4-48 Interface Configuration File Output
To configure the spanning tree mode to promiscuous, use this:
auto eth4 iface eth4 inet manual
up ip address add 0/0 dev $IFACE
up ip link set $IFACE up
up ip link set $IFACE promisc on
Alternatively, access the file /etc/network/interfaces/ifcfg-eth4 and type this:
This method provides persistent configuration settings for ifcfg-eth4.
Step 16. In the Virtual Machine Manager, select Show Virtual Hardware Details.