Introduction
This document describes steps required to deploy a sample environment for the Public Cloud Edge Interface (PCEI) Blueprint.
License
TBD
Deployment Architecture
The deployment architecture is shown below. All Addressing and naming shown are for example purposes only.
Deployment environment components:
- One Host Server
- One Edge Multi-Cluster Orchestrator VM
- Two Edge K8S Cluster VMs
- Internal network connectivity to the Host Server and to the VMs
Pre-Installation Requirements
Recommended Hardware Requirements
Requirement | Value |
---|---|
CPU | 18 Core / 36 vCPU |
RAM | 128 GB |
DISK | 500 GB |
NETWORK | 1Gbps NIC Port |
Network Requirements
- Internal Lab Network (RFC 1918 space)
- Internet access (NAT'd)
- IP routing in the Lab Network to reach VM interfaces
Software Prerequisites
Requirement | Value |
---|---|
Server OS | CentOS 7.x or above |
VM OS | Ubuntu 18.04 |
Upstream SW | Refer to Deployment Guide Section |
Installation High-Level Overview
The installation/deployment process consists of the following steps:
- Install CentOS on Host Server.
- Prepare Host Server for EMCO Deployment.
- Deploy EMCO.
Deployment Guide
Install CentOS on Host Server
- Connect to Host Server iLO interface.
- Start Virtual Console.
- Mount Virtual Media with CentOS 7 ISO.
- Install CentOS
- Assign correct IP address, Subnet, Gateway and DNS to the NIC.
- Include OpenSSH Server.
- Install KVM/virtualization.
- Add a user with admin privileges: onaplab user is used in this guide.
Prepare Host Server for EMCO Deployment
Step 1. Enable sudo without entering sudo password
sudo -i visudo # Uncomment the line below %wheel ALL=(ALL) NOPASSWD: ALL
Step 2. Add sudo user to wheel group:
usermod –aG wheel onaplab
Step 3. Enable nested virtualization
# Login as super user sudo -i # Check for the following parameter cat /sys/module/kvm_intel/parameters/nested N # If it is Y, there is nothing else to be done. It is already enabled! # If it is N, do the following: # Edit /etc/default/grub file, and include kvm-intel.nested=1 GRUB_CMDLINE_LINUX parameter GRUB_CMDLINE_LINUX="crashkernel=auto console=ttyS0,38400n8,kvm-intel.nested=1" # Rebuild your GRUB configuration sudo grub2-mkconfig -o /boot/grub2/grub.cfg # Enable nested KVM capabilities in /etc/modprobe.d/kvm.conf # By uncommenting the below line options kvm_intel nested=1 ### Reboot the server reboot # Now, we should have the nested KVM capabilities enabled cat /sys/module/kvm_intel/parameters/nested Y
Step 4. Install VNC Server
Follow instructions at:
https://www.tecmint.com/install-and-configure-vnc-server-in-centos-7/
Step 5. Modify libvirt bridge IP and route mode
This will allow connecting to VMs and pods directly from the Lab Network. Please replace the sample IPs with your IP addresses.
cat <<\EOF >> netdefault.xml <network> <name>default</name> <bridge name="vmbr0"/> <forward mode='route' dev='eno24'/> # <forward/> <ip address="10.121.7.129" netmask="255.255.255.224"> <dhcp> <range start="10.121.7.144" end="10.121.7.158"/> </dhcp> </ip> </network> EOF sudo virsh net-list sudo virsh net-destroy default sudo virsh net-undefine default sudo virsh net-define netdefault.xml sudo virsh net-start default sudo virsh net-autostart default
Add necessary routes to your Lab Network routers. The example below assumes the the Host Server IP address is 10.121.1.12:
ip route 10.121.7.128 255.255.255.224 10.121.1.12
Deploy EMCO
Step 1. Generate SSH Keys
# Run commands below on the Host Server ssh-keygen cd ~/.ssh chmod 600 id_rsa chmod 600 id_rsa.pub chmod 700 config chmod 600 known_hosts cat id_rsa.pub >> authorized_keys chmod 600 authorized_keys
Step 2. Download software and install EMCO
#### On the Host Server sudo yum install -y git deltarpm mkdir -p amcop_deploy cd amcop_deploy ## Download zip file from https://drive.google.com/file/d/1aFtChA454uAMuoEyV4m-jWiRDrZGyYgx/view?usp=sharing unzip amcop_install_v1.0.zip cd ~/amcop_deploy/aarna-stream/util-scripts ./prep_baremetal_centos.sh ####### Install EMCO/AMCOP # Edit inventory.ini file. Use IP address of Host Server and the username. cd ~/amcop_deploy/aarna-stream/anod_lite/ansible/deployment vi inventory.ini [deployment_host] 10.121.1.12 ansible_user=onaplab nohup ansible-playbook ./main.yml -i inventory.ini -e deployment_env=on-prem -e jump_host_user=onaplab --private-key=/home/onaplab/.ssh/id_rsa -e vm_user=onaplab &
Step 3. Monitor the installation
# On the Host Server cd /home/onaplab/aarna-stream/anod_lite/logs [onaplab@os12 logs]$ ls -l total 1980 -rw-r--r--. 1 root root 510417 Nov 24 07:06 cluster_setup.log -rw-r--r--. 1 root root 2019 Nov 24 06:54 create_vm.log -rw-r--r--. 1 root root 1366779 Nov 24 07:15 deploy_emco_components.log -rw-r--r--. 1 root root 138233 Nov 24 07:35 deploy_onap.log -rw-rw-r--. 1 onaplab onaplab 83 Nov 24 06:53 README.md tail -f create_vm.log tail -f cluster_setup.log tail -f deploy_emco_components.log tail -f deploy_onap.log
Deploy Edge Clusters
Step 1. Edit VM creation script.
# On the Host Server cd /home/onaplab/amcop_deploy/aarna-stream/util-scripts # Add "-cpu host" option to the end of the below line vi create_qem_vm.sh virt-install --connect qemu:///system --name $vm_name --ram $(($mem << 10)) --vcpus=$vCPU --os-type linux --os-variant $os_variant --disk path=/var/lib/libvirt/images/$vm_name/"$vm_name".qcow2,format=qcow2 --disk /var/lib/libvirt/images/$vm_name/$vm_name-cidata.iso,device=cdrom --import --network network=default --noautoconsole —cpu host # Save the file
Step 2. Deploy two Edge Cluster VMs.
These commands will create two Ubuntu 18.04 VMs with 100G Disk, 8 vcpu and 16G RAM and will copy the contents of the ~/.ssh/id_rsa.pub key file from the Host Server to the VMs' ~/.ssh/authorized_keys file.
sudo ./create_qem_vm.sh 2 edge_k8s-1 100 8 16 ubuntu18.04 $HOME/.ssh/id_rsa.pub onaplab sudo ./create_qem_vm.sh 2 edge_k8s-2 100 8 16 ubuntu18.04 $HOME/.ssh/id_rsa.pub onaplab
Step 3. Setup worker clusters inside VMs
# Find VM's IP addresses. On the Host Server run: [onaplab@os12 ~]$ sudo virsh list --all Id Name State ---------------------------------------------------- 6 amcop-vm-01 running 9 edge_k8s-1 running 10 edge_k8s-2 running [onaplab@os12 ~]$ sudo virsh domifaddr edge_k8s-1 Name MAC address Protocol Address ------------------------------------------------------------------------------- vnet1 52:54:00:19:96:72 ipv4 10.121.7.152/27 [onaplab@os12 ~]$ [onaplab@os12 ~]$ sudo virsh domifaddr edge_k8s-2 Name MAC address Protocol Address ------------------------------------------------------------------------------- vnet2 52:54:00:c0:47:8b ipv4 10.121.7.146/27 # ssh to each VM from the Host Server: ssh -i ~/.ssh/id_rsa.pub onaplab@10.121.7.152 ssh -i ~/.ssh/id_rsa.pub onaplab@10.121.7.146 # Perform the following tasks in each VM: sudo apt-get update -y sudo apt-get upgrade -y sudo apt-get install -y python-pip git clone https://git.onap.org/multicloud/k8s/ # Run script to setup KUD clusters nohup k8s/kud/hosting_providers/baremetal/aio.sh