PCEI R4 Installation Guide
- Oleg Berzin
Owned by Oleg Berzin
Introduction
This document describes steps required to deploy a sample environment for the Public Cloud Edge Interface (PCEI) Blueprint.
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 | 1 Gbps 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.
- Deploy Edge Clusters.
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 (Optional)
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. Please replace the interface name (eno24 used in the example) with you server's interface name.
cat <<\EOF >> netdefault.xml
<network>
<name>default</name>
<bridge name="vmbr0"/>
<forward mode='route' dev='eno24'/>
<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 echo "# Increase the server timeout value" >> ~/.ssh/config echo "ServerAliveInterval 120" >> ~/.ssh/config
Step 2. Download software and install EMCO
Note that the install process will:
- Deploy a VM amcop-vm-01
- Create a K8S cluster inside the VM
- Deploy EMCO components on the K8S cluster
- Deploy ONAP components on the K8S cluster
#### On the Host Server sudo yum install -y git deltarpm mkdir -p amcop_deploy cd amcop_deploy ## Download the installation package zip file wget --load-cookies /tmp/cookies.txt "https://docs.google.com/uc?export=download&confirm=$(wget --quiet --save-cookies /tmp/cookies.txt --keep-session-cookies --no-check-certificate 'https://docs.google.com/uc?export=download&id=1NE6dfbt3l6u2D6MRocdL04_xMjPn06HZ' -O- | sed -rn 's/.*confirm=([0-9A-Za-z_]+).*/\1\n/p')&id=1NE6dfbt3l6u2D6MRocdL04_xMjPn06HZ" -O amcop_install_v1.0.zip && rm -rf /tmp/cookies.txt unzip amcop_install_v1.0.zip sudo chown -R onaplab:onaplab ~/amcop_deploy/ cd ~/amcop_deploy/aarna-stream/util-scripts ./prep_baremetal_centos.sh ####### Install EMCO/AMCOP # Edit inventory.ini file. Use IP address of Host Server and the username. cd ~/amcop_deploy/aarna-stream/anod_lite/ansible/deployment vi inventory.ini [deployment_host] 10.121.1.12 ansible_user=onaplab nohup ansible-playbook ./main.yml -i inventory.ini -e deployment_env=on-prem -e jump_host_user=onaplab --private-key=/home/onaplab/.ssh/id_rsa -e vm_user=onaplab &
Step 3. Monitor the installation
# On the Host Server cd /home/onaplab/aarna-stream/anod_lite/logs [onaplab@os12 logs]$ 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
If Install fails and you need to restart, please do the cleanup steps below on the Host Server.
sudo virsh destroy amcop-vm-01 sudo virsh undefine amcop-vm-01 sudo virsh pool-destroy amcop-vm-01 sudo virsh pool-undefine amcop-vm-01 sudo rm /var/lib/libvirt/images/amcop-vm-01/amcop-vm-01-cidata.iso sudo rm /var/lib/libvirt/images/amcop-vm-01/amcop-vm-01.qcow2
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 onaplab@10.121.7.152 ssh 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 %
If the edge cluster deployment fails for any reason, please do the clean up steps below before you retry:
### Cleanup sudo virsh destroy edge_k8s-1 sudo virsh undefine edge_k8s-1 sudo virsh pool-destroy edge_k8s-1 sudo virsh pool-undefine edge_k8s-1 sudo rm /var/lib/libvirt/images/edge_k8s-1/edge_k8s-1-cidata.iso sudo rm /var/lib/libvirt/images/edge_k8s-1/edge_k8s-1.qcow2 sudo virsh destroy edge_k8s-2 sudo virsh undefine edge_k8s-2 sudo virsh pool-destroy edge_k8s-2 sudo virsh pool-undefine edge_k8s-2 sudo rm /var/lib/libvirt/images/edge_k8s-2/edge_k8s-2-cidata.iso sudo rm /var/lib/libvirt/images/edge_k8s-2/edge_k8s-2.qcow2
Modify sshd_config on VMs
To ensure that user onaplab can successfully ssh into EMCO and edge cluster VMs, add user onaplab to the sshd_config file.
## ssh to each VM: ssh onaplab@10.121.7.152 sudo -i cd /etc/ssh vi sshd_config AllowUsers ubuntu onaplab ## Save the changes and exit the file
Deployment Verification
EMCO Deployment Verification
Perform the following steps to verify correct EMCO deployment:
# Determine IP address of EMCO VM: [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 amcop-vm-01 Name MAC address Protocol Address ------------------------------------------------------------------------------- vnet0 52:54:00:1a:8e:8b ipv4 10.121.7.145/27 # ssh to EMCO VM. You should be able to ssh without specifying the key: [onaplab@os12 ~]$ ssh onaplab@10.121.7.145 # Verify K8S pods: onaplab@emco:~$ kubectl get pods --all-namespaces NAMESPACE NAME READY STATUS RESTARTS AGE kube-system calico-kube-controllers-6f954885fb-bb2mr 1/1 Running 0 28d kube-system calico-node-ldcpv 1/1 Running 0 28d kube-system coredns-6b968665c4-6558h 1/1 Running 0 28d kube-system coredns-6b968665c4-rq6bf 0/1 Pending 0 28d kube-system dns-autoscaler-5fc5fdbf6-njch7 1/1 Running 0 28d kube-system kube-apiserver-node1 1/1 Running 0 28d kube-system kube-controller-manager-node1 1/1 Running 0 28d kube-system kube-proxy-gg95d 1/1 Running 0 28d kube-system kube-scheduler-node1 1/1 Running 0 28d kube-system kubernetes-dashboard-6c7466966c-cjpxm 1/1 Running 0 28d kube-system nodelocaldns-7pxcs 1/1 Running 0 28d kube-system tiller-deploy-8756df4d9-zq52m 1/1 Running 0 28d onap dev-cassandra-0 1/1 Running 0 28d onap dev-cassandra-1 1/1 Running 0 28d onap dev-cassandra-2 1/1 Running 0 28d onap dev-cds-blueprints-processor-6d697cc4d6-wzlfj 0/1 Init:1/3 0 28d onap dev-cds-db-0 1/1 Running 0 28d onap dev-cds-py-executor-7dcdc5f7f6-tpfmg 1/1 Running 0 28d onap dev-cds-sdc-listener-f99d4588d-nt2tk 0/1 Init:0/1 4021 28d onap dev-cds-ui-7768bb4b-cfbzd 1/1 Running 0 28d onap dev-mariadb-galera-0 1/1 Running 0 28d onap dev-mariadb-galera-1 1/1 Running 0 28d onap dev-mariadb-galera-2 1/1 Running 0 28d onap4k8s clm-668c45d96d-99gpb 1/1 Running 0 28d onap4k8s emcoui-57846bd5df-c774f 1/1 Running 0 28d onap4k8s etcd-768d5b6cc-ptmmr 1/1 Running 0 28d onap4k8s middleend-6d67c9bf54-tvs7s 1/1 Running 0 28d onap4k8s mongo-7988cb488b-kf29q 1/1 Running 0 28d onap4k8s ncm-9f4b85787-nqnlm 1/1 Running 0 28d onap4k8s orchestrator-5fd4845f8f-qsxlf 1/1 Running 0 28d onap4k8s ovnaction-f794f65b6-w85ms 1/1 Running 0 28d onap4k8s rsync-7d9f5fbd9b-r72sp 1/1 Running 0 28d # Verify K8S services: onaplab@emco:~$ kubectl get svc --all-namespaces NAMESPACE NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE default kubernetes ClusterIP 10.233.0.1 <none> 443/TCP 28d kube-system coredns ClusterIP 10.233.0.3 <none> 53/UDP,53/TCP,9153/TCP 28d kube-system kubernetes-dashboard ClusterIP 10.233.35.188 <none> 443/TCP 28d kube-system tiller-deploy ClusterIP 10.233.33.249 <none> 44134/TCP 28d onap cassandra ClusterIP None <none> 7000/TCP,7001/TCP,7199/TCP,9042/TCP,9160/TCP,61621/TCP 28d onap cds-blueprints-processor-cluster ClusterIP 10.233.4.219 <none> 5701/TCP 28d onap cds-blueprints-processor-grpc ClusterIP 10.233.34.1 <none> 9111/TCP 28d onap cds-blueprints-processor-http ClusterIP 10.233.31.74 <none> 8080/TCP 28d onap cds-db ClusterIP None <none> 3306/TCP 28d onap cds-py-executor ClusterIP 10.233.43.240 <none> 50052/TCP,50053/TCP 28d onap cds-sdc-listener ClusterIP 10.233.2.48 <none> 8080/TCP 28d onap cds-ui NodePort 10.233.55.19 <none> 3000:30497/TCP 28d onap mariadb-galera ClusterIP None <none> 3306/TCP 28d onap4k8s clm NodePort 10.233.59.50 <none> 9061:31856/TCP 28d onap4k8s emcoui NodePort 10.233.2.5 <none> 9080:30480/TCP 28d onap4k8s etcd ClusterIP 10.233.54.80 <none> 2379/TCP,2380/TCP 28d onap4k8s middleend NodePort 10.233.11.225 <none> 9891:31289/TCP 28d onap4k8s mongo ClusterIP 10.233.19.133 <none> 27017/TCP 28d onap4k8s ncm NodePort 10.233.16.20 <none> 9031:32737/TCP 28d onap4k8s orchestrator NodePort 10.233.23.25 <none> 9015:31298/TCP 28d onap4k8s ovnaction NodePort 10.233.37.45 <none> 9053:32514/TCP,9051:31181/TCP 28d onap4k8s rsync NodePort 10.233.60.47 <none> 9041:30555/TCP 28d
Access EMCOUI GUI:
# Determine EMCOUI Service Port: onaplab@emco:~$ kubectl get svc emcoui -n onap4k8s NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE emcoui NodePort 10.233.2.5 <none> 9080:30480/TCP 28d # To connect to EMCOUI GUI use IP address of amcop-vm-01 and port 30480
To connect to EMCOUI GUI use IP address of amcop-vm-01 and port 30480:
Edge Cluster Deployment Verification
To verify deployment of Edge Clusters, perform the following steps:
# Determine Edge Cluster VM IP addresses: [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 ~]$ 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. You should be able to ssh without specifying the key: ssh onaplab@10.121.7.152 ssh onaplab@10.121.7.146 # Perform the following tasks inside the VMs: onaplab@localhost:~$ kubectl get pods --all-namespaces NAMESPACE NAME READY STATUS RESTARTS AGE kube-system cmk-rpgd7 2/2 Running 0 28d kube-system coredns-dff8fc7d-2xwrk 0/1 Pending 0 28d kube-system coredns-dff8fc7d-q2gcr 1/1 Running 0 28d kube-system dns-autoscaler-66498f5c5f-2kzmv 1/1 Running 0 28d kube-system kube-apiserver-localhost 1/1 Running 0 28d kube-system kube-controller-manager-localhost 1/1 Running 0 28d kube-system kube-flannel-8rm9p 1/1 Running 0 28d kube-system kube-multus-ds-amd64-mt9s5 1/1 Running 0 28d kube-system kube-proxy-ggk8m 1/1 Running 0 28d kube-system kube-scheduler-localhost 1/1 Running 0 28d kube-system kubernetes-dashboard-84999f8b5b-48xjq 1/1 Running 0 28d kube-system kubernetes-metrics-scraper-54fbb4d595-rw649 1/1 Running 0 28d kube-system local-volume-provisioner-bmkc6 1/1 Running 0 28d kube-system virtlet-vk7jl 3/3 Running 0 28d node-feature-discovery nfd-master-78nms 1/1 Running 0 28d node-feature-discovery nfd-worker-k4d5g 1/1 Running 45 28d operator nfn-agent-zlp9g 1/1 Running 0 28d operator nfn-operator-b768877d8-vcx7v 1/1 Running 0 28d operator ovn4nfv-cni-4c6rx 1/1 Running 0 28d # Verify connectivity to EMCO Cluster onaplab@localhost:~$ ping 10.121.7.145 PING 10.121.7.145 (10.121.7.145) 56(84) bytes of data. 64 bytes from 10.121.7.145: icmp_seq=1 ttl=64 time=0.457 ms 64 bytes from 10.121.7.145: icmp_seq=2 ttl=64 time=0.576 ms
Uninstall Guide
Perform the following steps to remove EMCO and Edge Clusters from the Host Server:
sudo virsh destroy amcop-vm-01 sudo virsh undefine amcop-vm-01 sudo virsh pool-destroy amcop-vm-01 sudo virsh pool-undefine amcop-vm-01 sudo rm /var/lib/libvirt/images/amcop-vm-01/amcop-vm-01-cidata.iso sudo rm /var/lib/libvirt/images/amcop-vm-01/amcop-vm-01.qcow2 sudo virsh destroy edge_k8s-1 sudo virsh undefine edge_k8s-1 sudo virsh pool-destroy edge_k8s-1 sudo virsh pool-undefine edge_k8s-1 sudo rm /var/lib/libvirt/images/edge_k8s-1/edge_k8s-1-cidata.iso sudo rm /var/lib/libvirt/images/edge_k8s-1/edge_k8s-1.qcow2 sudo virsh destroy edge_k8s-2 sudo virsh undefine edge_k8s-2 sudo virsh pool-destroy edge_k8s-2 sudo virsh pool-undefine edge_k8s-2 sudo rm /var/lib/libvirt/images/edge_k8s-2/edge_k8s-2-cidata.iso sudo rm /var/lib/libvirt/images/edge_k8s-2/edge_k8s-2.qcow2 sudo rm -rf ~/amcop_deploy sudo rm -rf ~/aarna_stream
License
References
AMCOP Quickstart Guide (Bare Metal) - EMCO Install Guide by Aarna Networks
AMCOP User Guide - EMCO Config Guide by Aarna Networks