Create Multi Node KVM Site on Equinix

• Log into the Equinix Console with your account credentials.

• Click Servers > Choose Deploy Type > Deploy On Demand Servers.

• Select a location and a server type.

• Select CentOS and choose CentOS 7 as the operating system version.

• Use the + button to set three servers and enter names for each of them.

• Prepare the user data for configuring packages. The following is a sample:

          #cloud-config

package_upgrade: true

packages:
    - qemu-kvm
    - qemu-img
    - virt-manager
    - libvirt
    - libvirt-python
    - libvirt-client
    - virt-install
    - virt-viewer
    - bridge-utils

runcmd:
   - systemctl start libvirtd
   - virsh net-undefine default
   - virsh net-destroy default
        

• Enable the Add User Data option and paste the configuration you prepared.

• Click the Deploy Now button to start deploying the servers.

After successful deployment, the Servers page displays the list of metal servers deployed.

After creating the servers, you are required to configure IP addresses by obtaining an elastic IP address block.

• Click IPs & Networks > IPs.

• Select Public IPv4, choose a location, choose a quantity, and then click Submit Request.

• Repeat for the Public IPv6 option.

• Verify the IP addresses are assigned in the PROJECT IP ADDRESSES tab.

The dual network interface controller (NIC) Mesh installation requires two VLANs, one for the internal interface and one for the external interface.

• Click IPs & Networks > Layer 2.

• Click Add New VLAN.

• Set the description to indicate it is an internal VLAN and click Add.

• Repeat the previous step to create another VLAN and set the description to indicate that it is an external VLAN.

• Verify the created VLANs.

• Go to the Servers page and click on a server for which the networking is required to be converted.

The server overview page displays the networking information:

• Select Network from the left menu to open the network page.

• Click Convert to Other Network Type.

• Select Hybrid in the network type selection window.

• Select Bonded in the options and select one of the VLANs to assign.

Note: You can select any of the created VLANs. The second VLAN will be assigned in the next step. Therefore, the order of configuration does not matter.

• Click ASSIGN NEW VLAN & CONVERT TO HYBRID NETWORKING.

• In the Network page, go to the Layer 2 section and click Add New Vlan.

• In the Add New Vlan window, select the other VLAN in the Network field and click Add.

Each metal instance requires two elastic IP addresses, one IP address for the network bridge interface and another for the Mesh external interface. To use an elastic IP address, it must be assigned to the metal instance.

After planning your address scheme, perform the following:

• From the Elastic IP block that was allocated in the previous section, assign two /32 addresses.

In this example, the allocated IP block is 147.28.142.240/29 and the first or last IP addresses are not used, as they represent the network and broadcast addresses.

The following table shows the sample IP addresses this example sets:

• In the Network page for your server, click Assign New Elastic in the Elastic IPs section.

• In the Add Elastic IP window, select Public IPv4 from the Address Type menu.

• Select an option from the Block menu.

• Select /32 from the Length menu.

• Select an address from the Address/Range menu.

• Click Add.

• Repeat the above steps to add another IP address.

• From the SERVER ACTIONS menu, select REBOOT to reboot and apply the changes.

• Repeat the steps above for all your bare metal instances.

• To SSH into the first metal instance, click ... > SSH Info.

• Click Copy to copy the SSH command displayed on the screen.

• Using your preferred SSH client, log into the server as root.

• Open the sysctl.conf file for editing:

          vi /etc/sysctl.conf
        

• Add the following contents to the end of the file and save the changes:

          vm.nr_hugepages = 1200
net.ipv4.ip_forward = 1
net.bridge.bridge-nf-call-ip6tables = 0
net.bridge.bridge-nf-call-iptables = 0
net.bridge.bridge-nf-call-arptables = 0
        

• Load the bridge filter module:

          modprobe br_netfilter
        

• Apply the changes:

          sysctl -p
        

• Copy the link to the latest software from the KVM Images page and use the wget command to download the image. The following is a sample command:

          wget -O /var/lib/libvirt/images/volterra.iso https://vesio.blob.core.windows.net/dev/images/centos/7.2009.10-202106210938/vsb-ves-ce-certifiedhw-generic-production-centos-7.2009.10-202106210938.1624275431.iso
        

• Wait for the download to complete. The download progress is indicated, and after completion, the command prompt is returned.

Enter the following command:

          qemu-img create /var/lib/libvirt/images/disk1.qcow2 30G
        

Define the internal and external KVM network interfaces and apply configuration parameters. The format of the device name is bond0.<VLAN ID>. In this example, the VLAN IDs assigned to the Distributed Cloud Services internal and external VLANs are 1000 and 1001, respectively.

• Create the internal VLAN interface file for editing:

          vi /etc/sysconfig/network-scripts/ifcfg-bond0.1000
        

• Add the following values and save the file:

          DEVICE=bond0.1000
ONBOOT=yes
BOOTPROTO=none
BRIDGE=br0
VLAN=yes
        

• Create the external VLAN interface:

          vi /etc/sysconfig/network-scripts/ifcfg-bond0.1001
        

• Add the following values and save the file:

          DEVICE=bond0.1001
ONBOOT=yes
BOOTPROTO=none
BRIDGE=br1
VLAN=yes
        

Define the Distributed Cloud Services internal VLAN interface network bridge IP addresses. The external Distributed Cloud Services interface use public IP addresses. For internal interface, the RFC1918 private addresses are used. This step configures only the bridge IP address for each instance.

Note: The internal IP address configuration is performed within F5® Distributed Cloud Console (Console) during a later step.

You can choose an IP address scheme per your choice for internal addressing, except the IP address space used by the bond0:0 interface. You can find the IP used by the bond0:0 interface by running the ip addr command.

In this example, the 192.168.100.0/24 address block is used. The following is the allocation scheme for this example:

Note: Bridge IP Address is the SiteLocalInside gateway and Distributed Cloud Services internal IP address is the SiteLocalInside IP prefix within Console.

• Create and open the internal bridge interface file for editing:

          vi /etc/sysconfig/network-scripts/ifcfg-br0
        

• Add DNS1, DNS2, IPADDR, and PREFIX settings and save the file. The following is a sample for the first metal instance:

          DEVICE=br0
STP=no
TYPE=Bridge
BOOTPROTO=none
DEFROUTE=yes
NAME=br0
ONBOOT=yes
DNS1=8.8.8.8
DNS2=8.8.4.4
IPADDR=192.168.100.10
PREFIX=24
        

• Repeat the steps for each metal instance with the appropriate IP address based on your address scheme.

Define the Distributed Cloud Services external VLAN network bridge IP addresses. We recommend you use the IP address pattern as shown in the example below. Ensure not to use the first and last IP addresses in the /29 address range.

This example uses the elastic IP block 147.28.142.240/29.

Note: This step configures only the bridge IP address for each instance. External IP address configuration is performed within Console during a later step.

The following is the scheme used for this example:

Note: Use the Elastic IP addresses from the block obtained in the Deploy Equinix Metal Servers chapter. The Bridge IP address is SiteLocal GW and the Distributed Cloud Services external IP address is the SiteLocal IP prefix within Console.

• Create and open the internal bridge interface file for editing:

          vi /etc/sysconfig/network-scripts/ifcfg-br1
        

• Add DNS1, DNS2, IPADDR, and PREFIX settings and save the file. The following is a sample for first metal instance:

          DEVICE=br1
STP=no
TYPE=Bridge
BOOTPROTO=none
DEFROUTE=yes
NAME=br1
ONBOOT=yes
DNS1=8.8.8.8
DNS2=8.8.4.4
IPADDR=147.28.142.241
PREFIX=29
        

• Repeat the steps for each metal instance with the appropriate IP address based on your address scheme.

• Reboot the metal instances and ping the elastic IP addresses to verify connectivity.

• Enter the following command:

          virt-install --name volterra \
--ram 16384 --vcpus=8 \
--os-variant=centos7.0 \
--network bridge=br1,model=virtio \
--network bridge=br0,model=virtio \
--accelerate \
--disk path=/var/lib/libvirt/images/disk1.qcow2,bus=virtio,cache=none,size=64 \
--cdrom /var/lib/libvirt/images/volterra.iso \
--noautoconsole --noreboot
        

• To start the VM, enter the following command:

          virsh start volterra
        

• To start the VM automatically when the underlying host OS is started, enter the following command:

          virsh autostart volterra
        

• After the VM is started, use the KVM console to connect to it. The ID of the VM assigned by the KVM is required for this.

• To obtain the ID assigned to the VM, enter the following command:

          virsh list
        

• Connect to the local console of the VM using the VM ID obtained in the previous step. The following is a sample command:

          virsh console 1
        

• At the login prompt, use admin as the username and Volterra123 as the password.

• After logging in for the first time, choose a new password. Passwords must be at least 8 characters and include at least one upper case letter.

• Use the tab key to open the configuration menu and navigate to configure-network, and then press the enter key to start network configuration.

• Use the arrow keys to navigate to the OUTSIDE option and press the space bar to select.

• On the screen that follows, populate the fields as shown in the following image:

• To confirm the changes, type y and then press the enter key.

• Use the tab key to open the configuration menu and navigate to configure-network and press enter key to start network configuration.

• Use the arrow keys to navigate to the INSIDE option and press the space bar to select.

• On the screen that follows, populate the fields as shown in the following image:

• Ensure that you set the primary interface as OUTSIDE.

• To confirm the changes, type y and then press the enter key.

To configure Mesh and register your site, a site token is required. You can create a new token or use an existing token. This example shows steps to create a new token.

• Log into Console.

• Click Multi-Cloud Network Connect.

• Click Manage > Site Management > Site Tokens.

• Click Add site token.

• Enter a name for the token and click Add site token.

• Note down the generated token code under the UID column for your token.

At this point, you can either connect using the KVM console or use SSH to connect. In case of SSH, you can use the elastic IP address you configured in previous section. This example shows connecting using SSH.

• Connect to the Mesh node using SSH and the elastic IP address for your instance.

• Use the tab key to open the configuration menu and navigate to configure and press the enter key to begin configuration.

• On the screen that follows, populate the fields as shown in the following image:

• For What is your token and What is your site name?, enter the same name and token for all nodes. Use the token created in the previous step.

• To confirm the changes, type y and then press the enter key. Your node will start to connect to Console for registration.

• Click Manage > Site Management > Registrations.

• Choose your site from the list of sites displayed under the Pending Registrations tab.

• Click the blue checkmark.

• On the page that opens, set the Cluster Size to 3. Make sure that you set each node to the same cluster size.

• Click Save and Exit to complete the registration process.

• Repeat the above steps for each of the remaining nodes.

Note: It may take a few minutes for the registration process to complete and the site to change to ONLINE.