In this guide we will perform an installation of Red Hat OpenShift Container Platform 4.9 on KVM Virtual Machines. OpenShift is a powerful, platform agnostic, enterprise-grade Kubernetes distribution focused on developer experience and application security. The project is developed and owned by Red Hat Software company. OpenShift Container Platform is built around containers orchestrated and managed by Kubernetes on a foundation of Red Hat Enterprise Linux.
The OpenShift platform offers automated installation, upgrades, and lifecycle management throughout the container stack – from the operating system, Kubernetes and cluster services, to deployed applications. Operating system that will be used on both the Control plan and Worker machines is Red Hat CoreOS (RHCOS). The RHCOS OS includes the kubelet, which is the Kubernetes node agent, and the CRI-O container runtime optimized for Kubernetes workloads.
In my installation the deployment is performed on a single node KVM compute server. This is not a production setup with high availability and should only be used for proof-of-concept and demo related purposes.
Red Hat’s recommendation on each cluster virtual machine minimum hardware requirements is as shown in the table below:Virtual Machine Operating System vCPU Virtual RAM Storage Bootstrap RHCOS 4 16 GB 120 GB Control plane RHCOS 4 16 GB 120 GB Compute RHCOS 2 8 GB 120 GB
But the preferred requirements for each cluster virtual machine are:Virtual Machine Operating System vCPU Virtual RAM Storage Bootstrap RHCOS 4 16 GB 120 GB Control plane RHCOS 8 16 GB 120 GB Compute RHCOS 6 8 GB 120 GB
The shared hardware requirements information for the virtual machines is not accurate since it depends on the workloads and desired cluster size when running in Production. Sizing can be done as deemed fit.
My Lab environment variables
- OpenShift 4 Cluster base domain: example.com ( to be substituted accordingly)
- OpenShift 4 Cluster name: ocp4 ( to be substituted accordingly)
- OpenShift KVM network bridge: openshift4
- OpenShift Network Block: 192.168.100.0/24
- OpenShift Network gateway address: 192.168.100.1
- Bastion / Helper node IP Address (Runs DHCP, Apache httpd, HAProxy, PXE, DNS) – 192.168.100.254
- NTP server used: time.google.com
Used Mac Addresses and IP Addresses:
| Machine Name | Mac Address (Generate yours and use) | DHCP Reserved IP Address |
| bootstrap.ocp4.example.com | 52:54:00:a4:db:5f | 192.168.100.10 |
| master01.ocp4.example.com | 52:54:00:8b:a1:17 | 192.168.100.11 |
| master02.ocp4.example.com | 52:54:00:ea:8b:9d | 192.168.100.12 |
| master03.ocp4.example.com | 52:54:00:f8:87:c7 | 192.168.100.13 |
| worker01.ocp4.example.com | 52:54:00:31:4a:39 | 192.168.100.21 |
| worker02.ocp4.example.com | 52:54:00:6a:37:32 | 192.168.100.22 |
| worker03.ocp4.example.com | 52:54:00:95:d4:ed | 192.168.100.23 |
Step 1: Setup KVM Infrastructure (On Hypervisor Node)
Install KVM in your hypervisor node using any of the guides in below links:
- Install KVM Hypervisor on Ubuntu
- How To Install KVM Hypervisor on Debian
- Install KVM on RHEL 8 / CentOS 8 / Rocky Linux
After installation verify your server CPU has support for Intel VT or AMD-V Virtualization extensions:
cat /proc/cpuinfo | egrep "vmx|svm"Creating Virtual Network (optional, you can use existing network)
Create a new virtual network configuration file
vim virt-net.xmlFile contents:
<network>
<name>openshift4</name>
<forward mode='nat'>
<nat>
<port start='1024' end='65535'/>
</nat>
</forward>
<bridge name='openshift4' stp='on' delay='0'/>
<domain name='openshift4'/>
<ip address='192.168.100.1' netmask='255.255.255.0'>
</ip>
</network>Create a virtual network using this file file created; modify if need be:
$ sudo virsh net-define --file virt-net.xml
Network openshift4 defined from virt-net.xmlSet the network to autostart on boot
$ sudo virsh net-autostart openshift4
Network openshift4 marked as autostarted
$ sudo virsh net-start openshift4
Network openshift4 startedConfirm that the bridge is available and active:
$ brctl show
bridge name bridge id STP enabled interfaces
openshift4 8000.5254002b479a yes
virbr0 8000.525400ad641d yesStep 2: Create Bastion / Helper Virtual Machine
Create a Virtual Machine that will host some key services from officially provided virt-builder images. The virtual machine will be used to run the following services:
- DNS Server (Bind)
- Apache httpd web server
- HAProxy Load balancer
- DHCP & PXE/TFTP services
- It will also be our bastion server for deploying and managing OpenShift platform (oc, openshift-install, kubectl, ansible)
Let’s first display available OS templates with command below:
$ virt-builder -lI’ll create a VM image from fedora-35 template; you can also choose a CentOS template(8 or 7):
sudo virt-builder fedora-35 --format qcow2 \
--size 20G -o /var/lib/libvirt/images/ocp-bastion-server.qcow2 \
--root-password password:StrongRootPassw0rdWhere:
- fedora-35 is the template used to create a new virtual machine
- /var/lib/libvirt/images/ocp-bastion-server.qcow2 is the path to VM qcow2 image
- StrongRootPassw0rd is the root user password
VM image creation progress will be visible in your screen
[ 1.0] Downloading: http://builder.libguestfs.org/fedora-35.xz
########################################################################################################################################################### 100.0%
[ 15.3] Planning how to build this image
[ 15.3] Uncompressing
[ 18.2] Resizing (using virt-resize) to expand the disk to 20.0G
[ 39.7] Opening the new disk
[ 44.1] Setting a random seed
[ 44.1] Setting passwords
[ 45.1] Finishing off
Output file: /var/lib/libvirt/images/ocp-bastion-server.qcow2
Output size: 20.0G
Output format: qcow2
Total usable space: 20.0G
Free space: 19.0G (94%)Now create a Virtual Machine to be used as DNS and DHCP server with virt-install
Using Linux bridge:
sudo virt-install \
--name ocp-bastion-server \
--ram 4096 \
--vcpus 2 \
--disk path=/var/lib/libvirt/images/ocp-bastion-server.qcow2 \
--os-type linux \
--os-variant rhel8.0 \
--network bridge=openshift4 \
--graphics none \
--serial pty \
--console pty \
--boot hd \
--importUsing openVSwitch bridge: Ref How To Use Open vSwitch Bridge on KVM Virtual Machines
sudo virt-install \
--name ocp-bastion-server \
--ram 4096 \
--disk path=/var/lib/libvirt/images/ocp-bastion-server.qcow2 \
--vcpus 2 \
--os-type linux \
--os-variant rhel8.0 \
--network=bridge:openshift4,model=virtio,virtualport_type=openvswitch \
--graphics none \
--serial pty \
--console pty \
--boot hd \
--importWhen your VM is created and running login as root user and password set initially:
Fedora 35 (Thirty Four)
Kernel 5.11.12-300.fc34.x86_64 on an x86_64 (ttyS0)
fedora login: root
Password: StrongRootPassw0rdYou can reset root password after installation if that’s your desired action:
[[email protected] ~]# passwd
Changing password for user root.
New password:
Retype new password:
passwd: all authentication tokens updated successfully.If the server didn’t get IP address from DHCP server you can set static IP manually on the primary interface:
# ip link show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP mode DEFAULT group default qlen 1000
link/ether 52:54:00:21:fb:33 brd ff:ff:ff:ff:ff:ff
# vi /etc/sysconfig/network-scripts/ifcfg-enp1s0
NAME="enp1s0" # Set network name, usually same as device name
DEVICE="enp1s0" # Set your interface name as shown while running ip link show command
ONBOOT="yes"
NETBOOT="yes"
BOOTPROTO="none"
TYPE="Ethernet"
PROXY_METHOD="none"
BROWSER_ONLY="no"
DEFROUTE="yes"
IPADDR=192.168.100.254 # Set your VM IP address
PREFIX=27 # Set Netmask Prefix
GATEWAY=192.168.100.1 # Set network gateway IP address
DNS1=8.8.8.8 # Set first DNS server to be used
DNS2=8.8.4.4 # Set secondary DNS server to be used
# Once configured bring up the interface using ifup command
# ifup enp1s0
Connection successfully activated (D-Bus active path: /org/freedesktop/NetworkManager/ActiveConnection/7)Test external connectivity from the VM:
# ping -c 2 8.8.8.8
PING 8.8.8.8 (8.8.8.8) 56(84) bytes of data.
64 bytes from 8.8.8.8: icmp_seq=1 ttl=117 time=4.98 ms
64 bytes from 8.8.8.8: icmp_seq=2 ttl=117 time=5.14 ms
--- 8.8.8.8 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1001ms
rtt min/avg/max/mdev = 4.981/5.061/5.142/0.080 ms
# ping -c 2 google.com
PING google.com (172.217.18.110) 56(84) bytes of data.
64 bytes from zrh04s05-in-f110.1e100.net (172.217.18.110): icmp_seq=1 ttl=118 time=4.97 ms
64 bytes from fra16s42-in-f14.1e100.net (172.217.18.110): icmp_seq=2 ttl=118 time=5.05 ms
--- google.com ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1002ms
rtt min/avg/max/mdev = 4.971/5.008/5.045/0.037 msPerform OS upgrade before deploying other services.
sudo dnf -y upgrade
sudo dnf -y install git vim wget curl bash-completion tree tar libselinux-python3 firewalldReboot the server after the upgrade is done.
sudo rebootConfirm you can access the VM through virsh console or ssh
$ sudo virsh list
Id Name State
-------------------------------------
1 ocp-bastion-server running
$ sudo virsh console ocp-bastion-server
Connected to domain 'ocp-bastion-server'
Escape character is ^] (Ctrl + ])
<ENTE>
fedora login:Enable domain autostart:
sudo virsh autostart ocp-bastion-serverStep 3: Install Ansible and Configure variables on Bastion / Helper node
Install Ansible configuration management tool on the Bastion machine
# Fedora
sudo dnf -y install git ansible vim wget curl bash-completion tree tar libselinux-python3
# CentOS 8 / Rocky Linux 8
sudo yum -y install epel-release
sudo yum -y install git ansible vim wget curl bash-completion tree tar libselinux-python3
# CentOS 7
sudo yum -y install epel-release
sudo yum -y install git ansible vim wget curl bash-completion tree tar libselinux-pythonWe have a Github repository with all the tasks and templates used in this guide. Clone the project to ~/ocp4_ansible directory.
cd ~/
git clone https://github.com/jmutai/ocp4_ansible.git
cd ~/ocp4_ansibleYou can view the directory structure using tree command:
$ tree
.
├── ansible.cfg
├── files
│ └── set-dns-serial.sh
├── handlers
│ └── main.yml
├── inventory
├── LICENSE
├── README.md
├── tasks
│ ├── configure_bind_dns.yml
│ ├── configure_dhcpd.yml
│ ├── configure_haproxy_lb.yml
│ └── configure_tftp_pxe.yml
├── templates
│ ├── default.j2
│ ├── dhcpd.conf.j2
│ ├── dhcpd-uefi.conf.j2
│ ├── haproxy.cfg.j2
│ ├── named.conf.j2
│ ├── pxe-bootstrap.j2
│ ├── pxe-master.j2
│ ├── pxe-worker.j2
│ ├── reverse.j2
│ └── zonefile.j2
└── vars
└── main.yml
5 directories, 21 filesEdit ansible configuration file and modify to suit your use.
$ vim ansible.cfg
[defaults]
inventory = inventory
command_warnings = False
filter_plugins = filter_plugins
host_key_checking = False
deprecation_warnings=False
retry_files = falseWhen not executing ansible as root user you can addprivilege_escalation section.
[privilege_escalation]
become = true
become_method = sudo
become_user = root
become_ask_pass = falseIf running on the localhost the inventory can be set as below:
$ vim inventory
[vmhost]
localhost ansible_connection=localThese are service handlers created and will be referenced in bastion setup process tasks.
$ vim handlers/main.yml
---
- name: restart tftp
service:
name: tftp
state: restarted
- name: restart bind
service:
name: named
state: restarted
- name: restart haproxy
service:
name: haproxy
state: restarted
- name: restart dhcpd
service:
name: dhcpd
state: restarted
- name: restart httpd
service:
name: httpd
state: restartedModify the default variables file inside vars folder:
vim vars/main.ymlDefine all the variables required correctly. Be careful not to have wrong values which will cause issues at the time of OpenShift installation.
---
ppc64le: false
uefi: false
disk: vda #disk where you are installing RHCOS on the masters/workers
helper:
name: "bastion" #hostname for your helper node
ipaddr: "192.168.100.254" #current IP address of the helper
networkifacename: "ens3" #interface of the helper node,ACTUAL name of the interface, NOT the NetworkManager name
dns:
domain: "example.com" #DNS server domain. Should match baseDomain inside the install-config.yaml file.
clusterid: "ocp4" #needs to match what you will for metadata.name inside the install-config.yaml file
forwarder1: "8.8.8.8" #DNS forwarder
forwarder2: "1.1.1.1" #second DNS forwarder
lb_ipaddr: "{{ helper.ipaddr }}" #Load balancer IP, it is optional, the default value is helper.ipaddr
dhcp:
router: "192.168.100.1" #default gateway of the network assigned to the masters/workers
bcast: "192.168.100.255" #broadcast address for your network
netmask: "255.255.255.0" #netmask that gets assigned to your masters/workers
poolstart: "192.168.100.10" #First address in your dhcp address pool
poolend: "192.168.100.50" #Last address in your dhcp address pool
ipid: "192.168.100.0" #ip network id for the range
netmaskid: "255.255.255.0" #networkmask id for the range.
ntp: "time.google.com" #ntp server address
dns: "" #domain name server, it is optional, the default value is set to helper.ipaddr
bootstrap:
name: "bootstrap" #hostname (WITHOUT the fqdn) of the bootstrap node
ipaddr: "192.168.100.10" #IP address that you want set for bootstrap node
macaddr: "52:54:00:a4:db:5f" #The mac address for dhcp reservation
masters:
- name: "master01" #hostname (WITHOUT the fqdn) of the master node (x of 3)
ipaddr: "192.168.100.11" #The IP address (x of 3) that you want set
macaddr: "52:54:00:8b:a1:17" #The mac address for dhcp reservation
- name: "master02"
ipaddr: "192.168.100.12"
macaddr: "52:54:00:ea:8b:9d"
- name: "master03"
ipaddr: "192.168.100.13"
macaddr: "52:54:00:f8:87:c7"
workers:
- name: "worker01" #hostname (WITHOUT the fqdn) of the worker node you want to set
ipaddr: "192.168.100.21" #The IP address that you want set (1st node)
macaddr: "52:54:00:31:4a:39" #The mac address for dhcp reservation (1st node)
- name: "worker02"
ipaddr: "192.168.100.22"
macaddr: "52:54:00:6a:37:32"
- name: "worker03"
ipaddr: "192.168.100.23"
macaddr: "52:54:00:95:d4:ed"Generating unique mac addresses for bootstrap, worker and master nodes
You can generate all required mac addresses using the command below:
date +%s | md5sum | head -c 6 | sed -e 's/\([0-9A-Fa-f]\{2\}\)/\1:/g' -e 's/\(.*\):$/\1/' | sed -e 's/^/52:54:00:/'Step 4: Install and Configure DHCP serveron Bastion / Helper node
Install dhcp-server rpm package using dnf or yum package manager.
sudo yum -y install dhcp-serverEnable dhcpd service to start on system boot
$ sudo systemctl enable dhcpd
Created symlink /etc/systemd/system/multi-user.target.wants/dhcpd.service → /usr/lib/systemd/system/dhcpd.service.Backup current dhcpd configuration file. If the server is not new you can modify existing configuration
sudo mv /etc/dhcp/dhcpd.conf /etc/dhcp/dhcpd.conf.bakTask to configure dhcp server on the bastion server:
$ vim tasks/configure_dhcpd.yml
---
# Setup OCP4 DHCP Server on Helper Node
- hosts: all
vars_files:
- ../vars/main.yml
handlers:
- import_tasks: ../handlers/main.yml
tasks:
- name: Write out dhcp file
template:
src: ../templates/dhcpd.conf.j2
dest: /etc/dhcp/dhcpd.conf
notify:
- restart dhcpd
when: not uefi
- name: Write out dhcp file (UEFI)
template:
src: ../templates/dhcpd-uefi.conf.j2
dest: /etc/dhcp/dhcpd.conf
notify:
- restart dhcpd
when: uefiConfigure DHCP server using ansible, defined variables and templates shared.
$ ansible-playbook tasks/configure_dhcpd.yml
PLAY [all] *******************************************************************************************************************************************************
TASK [Gathering Facts] *******************************************************************************************************************************************
ok: [localhost]
TASK [Write out dhcp file] ***************************************************************************************************************************************
changed: [localhost]
TASK [Write out dhcp file (UEFI)] ********************************************************************************************************************************
skipping: [localhost]
RUNNING HANDLER [restart dhcpd] **********************************************************************************************************************************
changed: [localhost]
PLAY RECAP *******************************************************************************************************************************************************
localhost : ok=3 changed=2 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0Confirm that dhcpd service is in running state:
$ systemctl status dhcpd
● dhcpd.service - DHCPv4 Server Daemon
Loaded: loaded (/usr/lib/systemd/system/dhcpd.service; enabled; vendor preset: disabled)
Active: active (running) since Tue 2021-08-17 19:35:06 EDT; 2min 42s ago
Docs: man:dhcpd(8)
man:dhcpd.conf(5)
Main PID: 24958 (dhcpd)
Status: "Dispatching packets..."
Tasks: 1 (limit: 4668)
Memory: 9.7M
CPU: 17ms
CGroup: /system.slice/dhcpd.service
└─24958 /usr/sbin/dhcpd -f -cf /etc/dhcp/dhcpd.conf -user dhcpd -group dhcpd --no-pid
...You can as well check generated configuration file:
$ cat /etc/dhcp/dhcpd.confStep 4: Configure OCP Zone on Bind DNS Serveron Bastion / Helper node
We can now begin the installation of DNS and DHCP server packages required to run OpenShift Container Platform on KVM.
sudo yum -y install bind bind-utilsEnable the service to start at system boot up
sudo systemctl enable namedInstall DNS Serialnumber generator script:
$ sudo vim /usr/local/bin/set-dns-serial.sh
#!/bin/bash
dnsserialfile=/usr/local/src/dnsserial-DO_NOT_DELETE_BEFORE_ASKING_CHRISTIAN.txt
zonefile=/var/named/zonefile.db
if [ -f zonefile ] ; then
echo $[ $(grep serial ${zonefile} | tr -d "\t"" ""\n" | cut -d';' -f 1) + 1 ] | tee ${dnsserialfile}
else
if [ ! -f ${dnsserialfile} ] || [ ! -s ${dnsserialfile} ]; then
echo $(date +%Y%m%d00) | tee ${dnsserialfile}
else
echo $[ $(< ${dnsserialfile}) + 1 ] | tee ${dnsserialfile}
fi
fi
##
##-30-Make the script executable:
sudo chmod a+x /usr/local/bin/set-dns-serial.shThis is the DNS Configuration task to be used:
$ vim tasks/configure_bind_dns.yml
---
# Configure OCP4 DNS Server on Helper Node
- hosts: all
vars_files:
- ../vars/main.yml
handlers:
- import_tasks: ../handlers/main.yml
tasks:
- name: Setup named configuration files
block:
- name: Write out named file
template:
src: ../templates/named.conf.j2
dest: /etc/named.conf
notify:
- restart bind
- name: Set zone serial number
shell: "/usr/local/bin/set-dns-serial.sh"
register: dymanicserialnumber
- name: Setting serial number as a fact
set_fact:
serialnumber: "{{ dymanicserialnumber.stdout }}"
- name: Write out "{{ dns.domain | lower }}" zone file
template:
src: ../templates/zonefile.j2
dest: /var/named/zonefile.db
mode: '0644'
notify:
- restart bind
- name: Write out reverse zone file
template:
src: ../templates/reverse.j2
dest: /var/named/reverse.db
mode: '0644'
notify:
- restart bindRun ansible playbook to configure bind dns server for OpenShift deployment.
$ ansible-playbook tasks/configure_bind_dns.yml
ansible-playbook tasks/configure_bind_dns.yml
PLAY [all] *******************************************************************************************************************************************************
TASK [Gathering Facts] *******************************************************************************************************************************************
ok: [localhost]
TASK [Write out named file] **************************************************************************************************************************************
changed: [localhost]
TASK [Set zone serial number] ************************************************************************************************************************************
changed: [localhost]
TASK [Setting serial number as a fact] ***************************************************************************************************************************
changed: [localhost]
TASK [Write out "example.com" zone file] **********************************************************************************************************************
changed: [localhost]
TASK [Write out reverse zone file] *******************************************************************************************************************************
changed: [localhost]
RUNNING HANDLER [restart bind] ***********************************************************************************************************************************
changed: [localhost]
PLAY RECAP *******************************************************************************************************************************************************
localhost : ok=7 changed=6 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0Forward DNS zone file is created under /var/named/zonefile.db and reverse DNS lookup file is /var/named/reverse.db
Check if the service is in running status:
$ systemctl status named
● named.service - Berkeley Internet Name Domain (DNS)
Loaded: loaded (/usr/lib/systemd/system/named.service; disabled; vendor preset: disabled)
Active: active (running) since Wed 2021-08-11 16:19:38 EDT; 4s ago
Process: 1340 ExecStartPre=/bin/bash -c if [ ! "$DISABLE_ZONE_CHECKING" == "yes" ]; then /usr/sbin/named-checkconf -z "$NAMEDCONF"; else echo "Checking of zo>
Process: 1342 ExecStart=/usr/sbin/named -u named -c ${NAMEDCONF} $OPTIONS (code=exited, status=0/SUCCESS)
Main PID: 1344 (named)
Tasks: 6 (limit: 4668)
Memory: 26.3M
CPU: 53ms
CGroup: /system.slice/named.service
└─1344 /usr/sbin/named -u named -c /etc/named.conf
Aug 11 16:19:38 fedora named[1344]: network unreachable resolving './NS/IN': 2001:500:1::53#53
Aug 11 16:19:38 fedora named[1344]: network unreachable resolving './NS/IN': 2001:500:200::b#53
Aug 11 16:19:38 fedora named[1344]: network unreachable resolving './NS/IN': 2001:500:9f::42#53
Aug 11 16:19:38 fedora named[1344]: network unreachable resolving './NS/IN': 2001:7fe::53#53
Aug 11 16:19:38 fedora named[1344]: network unreachable resolving './NS/IN': 2001:503:c27::2:30#53
Aug 11 16:19:38 fedora named[1344]: zone 1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa/IN: loaded serial 0
Aug 11 16:19:38 fedora named[1344]: all zones loaded
Aug 11 16:19:38 fedora named[1344]: managed-keys-zone: Initializing automatic trust anchor management for zone '.'; DNSKEY ID 20326 is now trusted, waiving the n>
Aug 11 16:19:38 fedora named[1344]: running
Aug 11 16:19:38 fedora systemd[1]: Started Berkeley Internet Name Domain (DNS).To test our DNS server we just execute:
$ dig @127.0.0.1 -t srv _etcd-server-ssl._tcp.ocp4.example.com
; <<>> DiG 9.16.19-RH <<>> @127.0.0.1 -t srv _etcd-server-ssl._tcp.ocp4.example.com
; (1 server found)
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 57264
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 3, AUTHORITY: 0, ADDITIONAL: 4
;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 1232
; COOKIE: e694eee032b927690100000061143bcf3df96ad3e49125d0 (good)
;; QUESTION SECTION:
;_etcd-server-ssl._tcp.ocp4.example.com. IN SRV
;; ANSWER SECTION:
_etcd-server-ssl._tcp.ocp4.example.com. 86400 IN SRV 0 10 2380 etcd-1.ocp4.example.com.
_etcd-server-ssl._tcp.ocp4.example.com. 86400 IN SRV 0 10 2380 etcd-2.ocp4.example.com.
_etcd-server-ssl._tcp.ocp4.example.com. 86400 IN SRV 0 10 2380 etcd-0.ocp4.example.com.
;; ADDITIONAL SECTION:
etcd-0.ocp4.example.com. 86400 IN A 192.168.100.11
etcd-1.ocp4.example.com. 86400 IN A 192.168.100.12
etcd-2.ocp4.example.com. 86400 IN A 192.168.100.13
;; Query time: 0 msec
;; SERVER: 127.0.0.1#53(127.0.0.1)
;; WHEN: Wed Aug 11 17:06:23 EDT 2021
;; MSG SIZE rcvd: 280As we’ve confirmed DNS server to be working as anticipated we can modify local system DNS server
$ nmcli connection show
NAME UUID TYPE DEVICE
enp1s0 c0ab6b8c-0eac-a1b4-1c47-efe4b2d1191f ethernet enp1s0
$ nmcli connection modify enp1s0 ipv4.dns "192.168.100.254"
$ nmcli connection reload
$ nmcli connection up enp1s0We can re-test if our Resolution is correct with:
$ host bootstrap.ocp4.example.com
bootstrap.ocp4.example.com has address 192.168.100.10Open firewall ports on the machine
sudo firewall-cmd --add-service={dhcp,tftp,http,https,dns} --permanent
sudo firewall-cmd --reloadStep 5: Setup TFTP Serviceon Bastion / Helper node
Install ftp related packages
sudo yum -y install tftp-server syslinuxAllow service in the firewall
sudo firewall-cmd --add-service=tftp --permanent
sudo firewall-cmd --reloadCreate TFTP Systemd unit file
$ sudo vim /etc/systemd/system/helper-tftp.service
[Unit]
Description=Starts TFTP on boot because of reasons
After=network.target
[Service]
Type=simple
ExecStart=/usr/local/bin/start-tftp.sh
TimeoutStartSec=0
Restart=always
RestartSec=30
[Install]
WantedBy=default.targetCreate TFTP Systemd helper script
sudo tee /usr/local/bin/start-tftp.sh<<EOF
#!/bin/bash
/usr/bin/systemctl start tftp > /dev/null 2>&1
##
##
EOFGive the script execution bits:
sudo chmod a+x /usr/local/bin/start-tftp.shReload Systemd daemon
sudo systemctl daemon-reloadStart tftp service
sudo systemctl enable --now tftp helper-tftpPopulate the default files for tftpboot
sudo mkdir -p /var/lib/tftpboot/pxelinux.cfgCopy syslinux files needed for PXE boot
sudo cp -rvf /usr/share/syslinux/* /var/lib/tftpbootCreate a directory for hosting the kernel and initramfs for PXE boot
sudo mkdir -p /var/lib/tftpboot/rhcosFiles to be downloaded
Obtain the RHEL kernel, initramfs, and rootfs files from the RHCOS image mirror page. The three main files to be downloaded:
- kernel:
rhcos-<version>-live-kernel-<architecture> - initramfs:
rhcos-<version>-live-initramfs.<architecture>.img - rootfs:
rhcos-<version>-live-rootfs.<architecture>.img
Download the CoreOS kernel file to this directory:
wget https://mirror.openshift.com/pub/openshift-v4/dependencies/rhcos/latest/latest/rhcos-installer-kernel-x86_64
sudo mv rhcos-installer-kernel-x86_64 /var/lib/tftpboot/rhcos/kernelThen the CoreOS Installer initramfs image:
wget https://mirror.openshift.com/pub/openshift-v4/dependencies/rhcos/latest/latest/rhcos-installer-initramfs.x86_64.img
sudo mv rhcos-installer-initramfs.x86_64.img /var/lib/tftpboot/rhcos/initramfs.imgNow we ned to relabel the files for selinux:
sudo restorecon -RFv /var/lib/tftpboot/rhcosList files in the directory:
$ ls /var/lib/tftpboot/rhcos
initramfs.img kernelApache httpd configurations
Install httpd server package
sudo yum -y install httpdWe need to change the configuration of the httpd from Listen on port 80 to Listen on Port 8080:
sudo vim /etc/httpd/conf/httpd.confSearch for the Line:
Listen 80Change the line to:
Listen 8080After that start httpd:
sudo systemctl enable httpd
sudo systemctl restart httpdOpen port 8080 in the firewall
sudo firewall-cmd --add-port=8080/tcp --permanent
sudo firewall-cmd --reloadCreate a directory in your web server root directory for CoreOS rootfs image
sudo mkdir -p /var/www/html/rhcosDownload Red Hat CoreOSrootfs image:
wget https://mirror.openshift.com/pub/openshift-v4/dependencies/rhcos/latest/latest/rhcos-live-rootfs.x86_64.imgMove the file to directory we created
sudo mv rhcos-live-rootfs.x86_64.img /var/www/html/rhcos/rootfs.img
sudo restorecon -RFv /var/www/html/rhcosYou can modify TFTP / PXE configuration task if need be:
$ vim tasks/configure_tftp_pxe.yml
---
# Configure OCP4 TFTP/PXE on Helper Node
- hosts: all
vars_files:
- ../vars/main.yml
handlers:
- import_tasks: ../handlers/main.yml
tasks:
- name: Set the bootstrap specific tftp file
template:
src: ../templates/pxe-bootstrap.j2
dest: "/var/lib/tftpboot/pxelinux.cfg/01-{{ bootstrap.macaddr | lower | regex_replace (':', '-')}}"
mode: 0555
notify:
- restart tftp
when: bootstrap is defined
- name: Set the master specific tftp files
template:
src: ../templates/pxe-master.j2
dest: "/var/lib/tftpboot/pxelinux.cfg/01-{{ item.macaddr | regex_replace (':', '-')}}"
mode: 0555
with_items: "{{ masters | lower }}"
notify:
- restart tftp
- name: Set the worker specific tftp files
template:
src: ../templates/pxe-worker.j2
dest: "/var/lib/tftpboot/pxelinux.cfg/01-{{ item.macaddr | regex_replace (':', '-')}}"
mode: 0555
with_items: "{{ workers | lower }}"
notify:
- restart tftp
when:
- workers is defined
- workers | length > 0Configure PXE environment for RHCOS using ansible
$ ansible-playbook tasks/configure_tftp_pxe.yml
PLAY [all] *******************************************************************************************************************************************************
TASK [Gathering Facts] *******************************************************************************************************************************************
ok: [localhost]
TASK [Set the bootstrap specific tftp file] **********************************************************************************************************************
changed: [localhost]
TASK [Set the master specific tftp files] ************************************************************************************************************************
changed: [localhost] => (item={'name': 'master01', 'ipaddr': '192.168.100.11', 'macaddr': '52:54:00:8b:a1:17'})
changed: [localhost] => (item={'name': 'master02', 'ipaddr': '192.168.100.12', 'macaddr': '52:54:00:ea:8b:9d'})
changed: [localhost] => (item={'name': 'master03', 'ipaddr': '192.168.100.13', 'macaddr': '52:54:00:f8:87:c7'})
TASK [Set the worker specific tftp files] ************************************************************************************************************************
changed: [localhost] => (item={'name': 'worker01', 'ipaddr': '192.168.100.21', 'macaddr': '52:54:00:31:4a:39'})
changed: [localhost] => (item={'name': 'worker02', 'ipaddr': '192.168.100.22', 'macaddr': '52:54:00:6a:37:32'})
changed: [localhost] => (item={'name': 'worker03', 'ipaddr': '192.168.100.23', 'macaddr': '52:54:00:95:d4:ed'})
RUNNING HANDLER [restart tftp] ***********************************************************************************************************************************
changed: [localhost]
PLAY RECAP *******************************************************************************************************************************************************
localhost : ok=5 changed=4 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0Headless environment considerations
With the consideration of the fact that we’re working in a headless environment, minimal setup of KVM without graphical interface. We need to ensure CoreOS booted VM will automatically choose the correct image and ignition file for the OS installation.
PXE Boot files are created inside the directory /var/lib/tftpboot/pxelinux.cfg
NOTE: Each of the file created should have a 01- before the MAC Address. See below example of bootstrap node.
Bootstrap node
Mac Address:
52:54:00:a4:db:5fThe file created will be
cat /var/lib/tftpboot/pxelinux.cfg/01-52-54-00-a4-db-5fWith contents:
default menu.c32
prompt 1
timeout 9
ONTIMEOUT 1
menu title ######## PXE Boot Menu ########
label 1
menu label ^1) Install Bootstrap Node
menu default
kernel rhcos/kernel
append initrd=rhcos/initramfs.img nomodeset rd.neednet=1 console=tty0 console=ttyS0 ip=dhcp coreos.inst=yes coreos.inst.install_dev=vda coreos.live.rootfs_url=http://192.168.100.254:8080/rhcos/rootfs.img coreos.inst.ignition_url=http://192.168.100.254:8080/ignition/bootstrap.ignMaster nodes
The file for each master has contents similar to this:
default menu.c32
prompt 1
timeout 9
ONTIMEOUT 1
menu title ######## PXE Boot Menu ########
label 1
menu label ^1) Install Master Node
menu default
kernel rhcos/kernel
append initrd=rhcos/initramfs.img nomodeset rd.neednet=1 console=tty0 console=ttyS0 ip=dhcp coreos.inst=yes coreos.inst.install_dev=vda coreos.live.rootfs_url=http://192.168.100.254:8080/rhcos/rootfs.img coreos.inst.ignition_url=http://192.168.100.254:8080/ignition/master.ignWorker nodes
The file for each worker node will looks similar to this:
default menu.c32
prompt 1
timeout 9
ONTIMEOUT 1
menu title ######## PXE Boot Menu ########
label 1
menu label ^1) Install Worker Node
menu default
kernel rhcos/kernel
append initrd=rhcos/initramfs.img nomodeset rd.neednet=1 console=tty0 console=ttyS0 ip=dhcp coreos.inst=yes coreos.inst.install_dev=vda coreos.live.rootfs_url=http://192.168.100.254:8080/rhcos/rootfs.img coreos.inst.ignition_url=http://192.168.100.254:8080/ignition/worker.ignYou can list all the files created using the following command:
$ ls -1 /var/lib/tftpboot/pxelinux.cfg
01-52:54:00:31:4a:39
01-52:54:00:6a:37:32
01-52:54:00:8b:a1:17
01-52:54:00:95:d4:ed
01-52:54:00:a4:db:5f
01-52:54:00:ea:8b:9d
01-52:54:00:f8:87:c7Step 6: Configure HAProxy as Load balanceron Bastion / Helper node
In this setup we’re using a software load balancer solution – HAProxy. In a Production setup of OpenShift Container Platform a hardware or highly available load balancer solution is required.
Install the package
sudo yum install -y haproxySet SEBool to allow haproxy connect any port:
sudo setsebool -P haproxy_connect_any 1Backup the default HAProxy configuration
sudo mv /etc/haproxy/haproxy.cfg /etc/haproxy/haproxy.cfg.defaultHere is HAProxy configuration ansible task:
$ vim tasks/configure_haproxy_lb.yml
---
# Configure OCP4 HAProxy Load balancer on Helper Node
- hosts: all
vars_files:
- ../vars/main.yml
tasks:
- name: Write out haproxy config file
template:
src: ../templates/haproxy.cfg.j2
dest: /etc/haproxy/haproxy.cfg
notify:
- restart haproxy
handlers:
- name: restart haproxy
ansible.builtin.service:
name: haproxy
state: restartedRun ansible-playbook using created task to configure HAProxy Load balancer for OpenShift
$ ansible-playbook tasks/configure_haproxy_lb.yml
PLAY [all] *******************************************************************************************************************************************************
TASK [Gathering Facts] *******************************************************************************************************************************************
ok: [localhost]
TASK [Write out haproxy config file] *****************************************************************************************************************************
changed: [localhost]
RUNNING HANDLER [restart haproxy] ********************************************************************************************************************************
changed: [localhost]
PLAY RECAP *******************************************************************************************************************************************************
localhost : ok=3 changed=2 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0Open the file for editing
sudo vim /etc/haproxy/haproxy.cfgConfiguration file is place in the file /etc/haproxy/haproxy.cfg
Configure SElinux for HAProxy to use the custom ports configured.
sudo semanage port -a 6443 -t http_port_t -p tcp
sudo semanage port -a 22623 -t http_port_t -p tcp
sudo semanage port -a 32700 -t http_port_t -p tcpOpen ports on the firewall
sudo firewall-cmd --add-service={http,https} --permanent
sudo firewall-cmd --add-port={6443,22623}/tcp --permanent
sudo firewall-cmd --reloadStep 7: Install OpenShift installer and CLI binaryon Bastion / Helper node
Download and install the OpenShift installer and client
OpenShift Client binary:
# Linux
wget https://mirror.openshift.com/pub/openshift-v4/clients/ocp/latest/openshift-client-linux.tar.gz
tar xvf openshift-client-linux.tar.gz
sudo mv oc kubectl /usr/local/bin
rm -f README.md LICENSE openshift-client-linux.tar.gz
# macOS
wget https://mirror.openshift.com/pub/openshift-v4/clients/ocp/latest/openshift-client-mac.tar.gz
tar xvf openshift-client-mac.tar.gz
sudo mv oc kubectl /usr/local/bin
rm -f README.md LICENSE openshift-client-mac.tar.gzOpenShift installer binary:
# Linux
wget https://mirror.openshift.com/pub/openshift-v4/clients/ocp/latest/openshift-install-linux.tar.gz
tar xvf openshift-install-linux.tar.gz
sudo mv openshift-install /usr/local/bin
rm -f README.md LICENSE openshift-install-linux.tar.gz
# macOS
wget https://mirror.openshift.com/pub/openshift-v4/clients/ocp/latest/openshift-install-mac.tar.gz
tar xvf openshift-install-mac.tar.gz
sudo mv openshift-install /usr/local/bin
rm -f README.md LICENSE openshift-install-mac.tar.gzCheck if you can run binaries:
$ openshift-install version
openshift-install 4.9.21
built from commit e9a66d5b21501c0a44f2f02cd245316798b0cdd9
release image quay.io/openshift-release-dev/[email protected]:fd96300600f9585e5847f5855ca14e2b3cafbce12aefe3b3f52c5da10c4476eb
release architecture amd64
$ oc version
Client Version: 4.9.21
$ kubectl version
Client Version: version.Info{Major:"1", Minor:"21", GitVersion:"v0.21.0-beta.1", GitCommit:"52a2fd9297c5a7221bed85db4cf303118bc396e3", GitTreeState:"clean", BuildDate:"2022-02-08T02:34:47Z", GoVersion:"go1.16.6", Compiler:"gc", Platform:"linux/amd64"}Create SSH Key Pairs
Now we need to create a SSH key pair to access to use later to access the CoreOS nodes
ssh-keygen -t rsa -N "" -f ~/.ssh/id_rsaStep 8: Generate ignition fileson Bastion / Helper node
We need to create the ignition files used for the installation of CoreOS machines
Download Pull Secret
We can store our pull secret in ~/.openshift directory:
mkdir ~/.openshiftVisit cloud.redhat.com and download your pull secret and save it under ~/.openshift/pull-secret
$ vim ~/.openshift/pull-secretCreate ocp4 directory
mkdir -p ~/ocp4
cd ~/We can now create OpenShift installation yaml file install-config-base.yaml:
cat <<EOF > install-config-base.yaml
apiVersion: v1
baseDomain: example.com
compute:
- hyperthreading: Enabled
name: worker
replicas: 0
controlPlane:
hyperthreading: Enabled
name: master
replicas: 3
metadata:
name: ocp4
networking:
clusterNetworks:
- cidr: 10.254.0.0/16
hostPrefix: 24
networkType: OpenShiftSDN
serviceNetwork:
- 172.30.0.0/16
platform:
none: {}
pullSecret: '$(< ~/.openshift/pull-secret)'
sshKey: '$(< ~/.ssh/id_rsa.pub)'
EOFYou can further modify the contents accordingly:
$ vim install-config-base.yaml
apiVersion: v1
baseDomain: example.com
compute:
- hyperthreading: Enabled
name: worker
replicas: 0
controlPlane:
hyperthreading: Enabled
name: master
replicas: 3
metadata:
name: ocp4
networking:
clusterNetworks:
- cidr: 10.128.0.0/14
hostPrefix: 23
networkType: OpenShiftSDN
serviceNetwork:
- 172.30.0.0/16
platform:
none: {}
pullSecret: 'paste-as-obtained-from-https://cloud.redhat.com'
sshKey: 'PASTE-SSH-PUBLIC-KEY'Copy the install-config-base.yaml file into the ocp4 directory with the name install-config.yaml
cd ~/
cp install-config-base.yaml ocp4/install-config.yamlChange into ocp4 directory
cd ocp4The directory must be empty every time you need to generate ignition files.
To create the Kubernetes manifest files run:
$ openshift-install create manifests
$ ls
manifests openshift
# All files
$ tree
.
├── manifests
│ ├── 04-openshift-machine-config-operator.yaml
│ ├── cluster-config.yaml
│ ├── cluster-dns-02-config.yml
│ ├── cluster-infrastructure-02-config.yml
│ ├── cluster-ingress-02-config.yml
│ ├── cluster-network-01-crd.yml
│ ├── cluster-network-02-config.yml
│ ├── cluster-proxy-01-config.yaml
│ ├── cluster-scheduler-02-config.yml
│ ├── cvo-overrides.yaml
│ ├── kube-cloud-config.yaml
│ ├── kube-system-configmap-root-ca.yaml
│ ├── machine-config-server-tls-secret.yaml
│ ├── openshift-config-secret-pull-secret.yaml
│ └── openshift-kubevirt-infra-namespace.yaml
└── openshift
├── 99_kubeadmin-password-secret.yaml
├── 99_openshift-cluster-api_master-user-data-secret.yaml
├── 99_openshift-cluster-api_worker-user-data-secret.yaml
├── 99_openshift-machineconfig_99-master-ssh.yaml
├── 99_openshift-machineconfig_99-worker-ssh.yaml
└── openshift-install-manifests.yaml
2 directories, 21 filesDisable pods scheduling on master nodes by changing mastersSchedulable parameter value from true to false
sed -i 's/true/false/' manifests/cluster-scheduler-02-config.ymlNow create the ignition files:
$ openshift-install create ignition-configs
INFO Consuming Common Manifests from target directory
INFO Consuming Openshift Manifests from target directory
INFO Consuming Master Machines from target directory
INFO Consuming Worker Machines from target directory
INFO Consuming OpenShift Install (Manifests) from target directory
INFO Ignition-Configs created in: . and authDirectory contents after generation of ignition files is as seen below
$ ls
auth bootstrap.ign master.ign metadata.json worker.ign
$ tree
.
├── auth
│ ├── kubeadmin-password
│ └── kubeconfig
├── bootstrap.ign
├── master.ign
├── metadata.json
└── worker.ign
1 directory, 6 filesCopy the files to our httpd server /var/www/html inside ocp4 directory we created earlier on
sudo mkdir -p /var/www/html/ignition
sudo cp -v *.ign /var/www/html/ignition
sudo chmod 644 /var/www/html/ignition/*.ign
sudo restorecon -RFv /var/www/html/Confirm files were copied
$ ls /var/www/html/ignition/
bootstrap.ign master.ign worker.ignEnsure all services are enabled and running
sudo systemctl enable --now haproxy.service dhcpd httpd tftp named
sudo systemctl restart haproxy.service dhcpd httpd tftp named
sudo systemctl status haproxy.service dhcpd httpd tftp namedHAProxy service status
$ systemctl status haproxy
● haproxy.service - HAProxy Load Balancer
Loaded: loaded (/usr/lib/systemd/system/haproxy.service; enabled; vendor preset: disabled)
Active: active (running) since Wed 2021-08-11 20:05:40 EDT; 44s ago
Process: 3129 ExecStartPre=/usr/sbin/haproxy -f $CONFIG -c -q $OPTIONS (code=exited, status=0/SUCCESS)
Main PID: 3137 (haproxy)
Tasks: 3 (limit: 4668)
Memory: 34.6M
CPU: 78ms
CGroup: /system.slice/haproxy.service
├─3137 /usr/sbin/haproxy -Ws -f /etc/haproxy/haproxy.cfg -p /run/haproxy.pid
└─3140 /usr/sbin/haproxy -Ws -f /etc/haproxy/haproxy.cfg -p /run/haproxy.pidStep 9: Create Bootstrap, Masters and Worker VMs(On Hypervisor Node)
Start with the creation of Bootstrap Virtual Machine.
sudo virt-install -n bootstrap.ocp4.example.com \
--description "Bootstrap Machine for Openshift 4 Cluster" \
--ram=8192 \
--vcpus=4 \
--os-type=Linux \
--os-variant=rhel8.0 \
--noreboot \
--disk pool=default,bus=virtio,size=50 \
--graphics none \
--serial pty \
--console pty \
--pxe \
--network bridge=openshift4,mac=52:54:00:a4:db:5fTo check if there are any errors in PXE boot process use the following command on the Bastion machine:
$ journalctl -f
# Or for specific service
$ journalctl -f -u tftp
$ journalctl -f -u dhcpdA successful instance creation will give you an output on the console – with domain creation completion message and how to start it.
[ 50.120576] [1226]: Remounting '/etc' read-only in with options 'seclabel,attr2,discard,inode64,logbufs=8,logbsize=32k,noquota'.
[ 50.125181] [1227]: Unmounting '/etc'.
[ 50.135577] [1228]: Remounting '/var' read-only in with options 'seclabel,attr2,discard,inode64,logbufs=8,logbsize=32k,noquota'.
[ 50.138117] [1229]: Unmounting '/var'.
[ 50.151944] XFS (loop0): Unmounting Filesystem
[ 50.165537] systemd-shutdown[1]: All filesystems unmounted.
[ 50.166730] systemd-shutdown[1]: Deactivating swaps.
[ 50.167793] systemd-shutdown[1]: All swaps deactivated.
[ 50.168888] systemd-shutdown[1]: Detaching loop devices.
[ 50.170315] systemd-shutdown[1]: Not all loop devices detached, 1 left.
[ 50.177714] kvm: exiting hardware virtualization
[ 50.192437] reboot: Restarting system
[ 50.193011] reboot: machine restart
Domain creation completed.
You can restart your domain by running:
virsh --connect qemu:///system start bootstrap.ocp4.example.comStart bootstrap node domain:
sudo virsh --connect qemu:///system start bootstrap.ocp4.example.comCreation of the three Master nodes, set correct VM name, network, and mac address.
# Create Master01 VM
sudo virt-install -n master01.ocp4.example.com \
--description "Master01 Machine for Openshift 4 Cluster" \
--ram=8192 \
--vcpus=4 \
--os-type=Linux \
--os-variant=rhel8.0 \
--noreboot \
--disk pool=default,bus=virtio,size=50 \
--graphics none \
--serial pty \
--console pty \
--pxe \
--network bridge=openshift4,mac=52:54:00:8b:a1:17
# Create Master02 VM
sudo virt-install -n master02.ocp4.example.com \
--description "Master02 Machine for Openshift 4 Cluster" \
--ram=8192 \
--vcpus=4 \
--os-type=Linux \
--os-variant=rhel8.0 \
--noreboot \
--disk pool=default,bus=virtio,size=50 \
--graphics none \
--serial pty \
--console pty \
--pxe \
--network bridge=openshift4,mac=52:54:00:ea:8b:9d
# Create Master03 VM
sudo virt-install -n master03.ocp4.example.com \
--description "Master03 Machine for Openshift 4 Cluster" \
--ram=8192 \
--vcpus=4 \
--os-type=Linux \
--os-variant=rhel8.0 \
--noreboot \
--disk pool=default,bus=virtio,size=50 \
--graphics none \
--serial pty \
--console pty \
--pxe \
--network bridge=openshift4,mac=52:54:00:f8:87:c7Start Master nodes domains
$ sudo virsh --connect qemu:///system start master01.ocp4.example.com
Domain 'master01.ocp.example.com' started
$ sudo virsh --connect qemu:///system start master02.ocp4.example.com
Domain 'master02.ocp.example.com' started
$ sudo virsh --connect qemu:///system start master03.ocp4.example.com
Domain 'master03.ocp.example.com' startedWorker nodes installation using virt-install and PXE boot.
# Create Worker01 VM
sudo virt-install -n worker01.ocp4.example.com \
--description "Worker01 Machine for Openshift 4 Cluster" \
--ram=8192 \
--vcpus=4 \
--os-type=Linux \
--os-variant=rhel8.0 \
--noreboot \
--disk pool=default,bus=virtio,size=50 \
--graphics none \
--serial pty \
--console pty \
--pxe \
--network bridge=openshift4,mac=52:54:00:31:4a:39
# Create Worker02 VM
sudo virt-install -n worker02.ocp4.example.com \
--description "Worker02 Machine for Openshift 4 Cluster" \
--ram=8192 \
--vcpus=4 \
--os-type=Linux \
--os-variant=rhel8.0 \
--noreboot \
--disk pool=default,bus=virtio,size=50 \
--graphics none \
--serial pty \
--console pty \
--pxe \
--network bridge=openshift4,mac=52:54:00:6a:37:32
# Create Worker03 VM
sudo virt-install -n worker03.ocp4.example.com \
--description "Worker03 Machine for Openshift 4 Cluster" \
--ram=8192 \
--vcpus=4 \
--os-type=Linux \
--os-variant=rhel8.0 \
--noreboot \
--disk pool=default,bus=virtio,size=50 \
--graphics none \
--serial pty \
--console pty \
--pxe \
--network bridge=openshift4,mac=52:54:00:95:d4:edStart Worker machine domains
$ sudo virsh --connect qemu:///system start worker01.ocp4.example.com
Domain 'worker01.ocp.example.com' started
$ sudo virsh --connect qemu:///system start worker02.ocp4.example.com
Domain 'worker02.ocp.example.com' started
$ sudo virsh --connect qemu:///system start worker03.ocp4.example.com
Domain 'worker03.ocp.example.com' startedOnce the master nodes are up and working the logs on the bootstrap node will show Succeeded
Aug 12 02:45:09 bootstrap.ocp.example.com bootkube.sh[61387]: Tearing down temporary bootstrap control plane...
Aug 12 02:45:10 bootstrap.ocp.example.com bootkube.sh[61387]: Sending bootstrap-finished event.Waiting for CEO to finish...
Aug 12 02:45:11 bootstrap.ocp.example.com bootkube.sh[61387]: W0812 02:45:11.179220 1 etcd_env.go:287] cipher is not supported for use with etcd: "TLS_ECDHE_ECDSA_WITH_CHACHA20_POLY1305_SHA256"
Aug 12 02:45:11 bootstrap.ocp.example.com bootkube.sh[61387]: W0812 02:45:11.179352 1 etcd_env.go:287] cipher is not supported for use with etcd: "TLS_ECDHE_RSA_WITH_CHACHA20_POLY1305_SHA256"
Aug 12 02:45:11 bootstrap.ocp.example.com bootkube.sh[61387]: I0812 02:45:11.285760 1 waitforceo.go:64] Cluster etcd operator bootstrapped successfully
Aug 12 02:45:11 bootstrap.ocp.example.com bootkube.sh[61387]: I0812 02:45:11.287327 1 waitforceo.go:58] cluster-etcd-operator bootstrap etcd
Aug 12 02:45:11 bootstrap.ocp.example.com bootkube.sh[61387]: bootkube.service complete
Aug 12 02:45:11 bootstrap.ocp.example.com systemd[1]: bootkube.service: Succeeded.Ensure all domains are set to autostart:
for i in {1..3}; do
sudo virsh autostart master0${i}.ocp4.example.com
sudo virsh autostart worker0${i}.ocp4.example.com
doneValidate the settings
$ virsh list --autostart
Id Name State
----------------------------------------------
53 master01.ocp4.example.com running
55 master02.ocp4.example.com running
57 master03.ocp4.example.com running
59 worker01.ocp4.example.com running
61 worker03.ocp4.example.com running
63 worker02.ocp4.example.com runningStep 10: Login to OpenShift CLI / Web Consoleon Bastion / Helper node
First, login to your cluster using generated kubeconfig file.
export KUBECONFIG=/root/ocp4/auth/kubeconfigOr copy kubeconfig file to ~/.kube directory to make it default:
mkdir ~/.kube
sudo cp /root/ocp4/auth/kubeconfig ~/.kube/config
sudo chown $USER ~/.kube/configEnable bash completion for oc and kubectl commands
$ vim ~/.bashrc
source <(oc completion bash)
source <(kubectl completion bash)
# Source bashrc file
$ source ~/.bashrcRun the following command to confirm who you’re logged into cluster as
$ oc whoami
system:adminPrinting API Endpoint URL:
$ oc whoami --show-server
https://api.ocp4.example.com:6443Confirm the version of OpenShift deployed in your KVM powered infrastructure:
$ oc get clusterversions.config.openshift.io
NAME VERSION AVAILABLE PROGRESSING SINCE STATUS
version False True 25m Working towards 4.9.17: 654 of 676 done (96% complete)
# Some minutes later
$ oc get clusterversions.config.openshift.io
NAME VERSION AVAILABLE PROGRESSING SINCE STATUS
version 4.9.17 True False 43m Cluster version is 4.9.17Check available nodes and their current status:
# oc get nodes
NAME STATUS ROLES AGE VERSION
master01.ocp.example.com Ready master 13m v1.22.3+e790d7f
master02.ocp.example.com Ready master 13m v1.22.3+e790d7f
master03.ocp.example.com Ready master 12m v1.22.3+e790d7fYour install may be waiting for worker nodes to get approved. Normally the machineconfig node approval operator takes care of this for you. However, sometimes this needs to be done manually. Check pending CSRs with the following command.
$ oc get csr
NAME AGE SIGNERNAME REQUESTOR CONDITION
csr-9cwd2 18m kubernetes.io/kube-apiserver-client-kubelet system:serviceaccount:openshift-machine-config-operator:node-bootstrapper Approved,Issued
csr-cqqvx 18m kubernetes.io/kubelet-serving system:node:master01.ocp.example.com Approved,Issued
csr-hdr7t 6m15s kubernetes.io/kube-apiserver-client-kubelet system:serviceaccount:openshift-machine-config-operator:node-bootstrapper Pending
csr-lcjsl 18m kubernetes.io/kubelet-serving system:node:master02.ocp.example.com Approved,Issued
csr-p9nj8 17m kubernetes.io/kubelet-serving system:node:master03.ocp.example.com Approved,Issued
csr-qkbrd 19m kubernetes.io/kube-apiserver-client-kubelet system:serviceaccount:openshift-machine-config-operator:node-bootstrapper Approved,Issued
csr-sxwlz 5m57s kubernetes.io/kube-apiserver-client-kubelet system:serviceaccount:openshift-machine-config-operator:node-bootstrapper Pending
csr-v244r 18m kubernetes.io/kube-apiserver-client-kubelet system:serviceaccount:openshift-machine-config-operator:node-bootstrapper Approved,Issued
system:openshift:openshift-authenticator 17m kubernetes.io/kube-apiserver-client system:serviceaccount:openshift-authentication-operator:authentication-operator Approved,IssuedYou can approve all pending CSRs with the following command:
oc get csr -o go-template='{{range .items}}{{if not .status}}{{.metadata.name}}{{"\n"}}{{end}}{{end}}' | xargs oc adm certificate approveList cluster nodes after csr approval
$ oc get nodes
NAME STATUS ROLES AGE VERSION
master01.ocp4.example.com Ready master 22m v1.22.3+e790d7f
master02.ocp4.example.com Ready master 22m v1.22.3+e790d7f
master03.ocp4.example.com Ready master 21m v1.22.3+e790d7f
worker01.ocp4.example.com Ready worker 2m2s v1.22.3+e790d7f
worker02.ocp4.example.com Ready worker 2m4s v1.22.3+e790d7fTo access a worker or master node shell use either of below methods:
# SSH
$ ssh [email protected]master01.ocp4.example.com
# using oc debug
$ oc debug node/master01.ocp4.example.com
Starting pod/master01ocp4examplecom-debug ...
To use host binaries, run `chroot /host`
chroot /host
Pod IP: 192.168.100.11
If you don't see a command prompt, try pressing enter.
sh-4.4#
sh-4.4# chroot /host
sh-4.4# bash
[[email protected] /]#Check the status of all Cluster operators. All should return [true in AVAILABLE] and [false in DEGRADED] states.
$ oc get co
NAME VERSION AVAILABLE PROGRESSING DEGRADED SINCE
authentication 4.9.17 True False False 44m
baremetal 4.9.17 True False False 76m
cloud-credential 4.9.17 True False False 81m
cluster-autoscaler 4.9.17 True False False 76m
config-operator 4.9.17 True False False 77m
console 4.9.17 True False False 50m
csi-snapshot-controller 4.9.17 True False False 76m
dns 4.9.17 True False False 76m
etcd 4.9.17 True False False 70m
image-registry 4.9.17 True False False 66m
ingress 4.9.17 True False False 53m
insights 4.9.17 True False False 71m
kube-apiserver 4.9.17 True False False 68m
kube-controller-manager 4.9.17 True False False 75m
kube-scheduler 4.9.17 True False False 75m
kube-storage-version-migrator 4.9.17 True False False 77m
machine-api 4.9.17 True False False 76m
machine-approver 4.9.17 True False False 76m
machine-config 4.9.17 True False False 75m
marketplace 4.9.17 True False False 75m
monitoring 4.9.17 True False False 52m
network 4.9.17 True False False 78m
node-tuning 4.9.17 True False False 76m
openshift-apiserver 4.9.17 True False False 65m
openshift-controller-manager 4.9.17 True False False 76m
openshift-samples 4.9.17 True False False 65m
operator-lifecycle-manager 4.9.17 True False False 76m
operator-lifecycle-manager-catalog 4.9.17 True False False 76m
operator-lifecycle-manager-packageserver 4.9.17 True False False 65m
service-ca 4.9.17 True False False 77m
storage 4.9.17 True False False 77mDeployed applications are exposed through ingress routes on the base domain apps.ocp4.example.com:
$ oc get ingresscontroller default -n openshift-ingress-operator -o jsonpath='{.status.domain}'
apps.ocp4.example.comLogin to the web console
To access OpenShift Web management console get login URL
$ oc whoami --show-console
https://console-openshift-console.apps.ocp4.example.comDefault login credentials are stored in the file ocp4/auth/kubeadmin-password
$ cat ocp4/auth/kubeadmin-passwordUsername is kubeadmin and password is as stored in the file.
Change your OpenShift updates channel to fast under Administration > Cluster Settings
Click on Channel to update
Upgrading to latest release of OpenShift
If you didn’t install the latest release, then just run the following to upgrade.
oc adm upgrade --to-latestConfiguring Csr automatic approval with systemd timer (Optional)
Copy kubeconfig file to a /etc directory to avoid permission issues for root user home directory:
sudo cp /root/ocp4/auth/kubeconfig /etc/ocp_kubeconfig
sudo chmod 0644 /etc/ocp_kubeconfigCreate a bash script that can be used to approve pending CSRs:
$ sudo vim /usr/local/bin/approve_csr.sh
#!/bin/bash
export KUBECONFIG=/etc/ocp_kubeconfig
oc get csr -o go-template='{{range .items}}{{if not .status}}{{.metadata.name}}{{"\n"}}{{end}}{{end}}' | xargs oc adm certificate approveMake the script executable
sudo chmod a+x /usr/local/bin/approve_csr.shCreate CSR approval service:
$ sudo vim /etc/systemd/system/ocp_csr_approval.service
# This service unit is for approving pending csr in OpenShift 4 cluster
# By Josphat Mutai
# Licensed under GPL V2
#
[Unit]
Description=Approve pending csr in OpenShift 4 cluster
Wants=ocp_csr_approval.timer
[Service]
Type=oneshot
ExecStart=/usr/local/bin/approve_csr.sh
[Install]
WantedBy=multi-user.targetNext we create the timer unit file inside /etc/systemd/system directory:
$ sudo vim /etc/systemd/system/ocp_csr_approval.timer
# This timer unit is for approving pending csr in OpenShift 4 cluster
# By Josphat Mutai
# Licensed under GPL V2
#
[Unit]
Description=Approve pending csr in OpenShift 4 cluster
Requires=ocp_csr_approval.service
[Timer]
Unit=ocp_csr_approval.service
OnCalendar=*-*-* *:*:00
[Install]
WantedBy=timers.targetThe OnCalendar time specification *-*-* *:*:00 should trigger the timer to execute the ocp_csr_approval.service unit every minute
Reload systemd units
sudo systemctl daemon-reloadBefore we install the timer, we can first test the service.
sudo systemctl start ocp_csr_approval.serviceEnable timer unit if test was successful.
sudo systemctl enable --now ocp_csr_approval.timerCheck execution status
journalctl -S today -f -u ocp_csr_approval.serviceOur Systemd timer should be listed with other systemd timers:
$ systemctl list-timers
NEXT LEFT LAST PASSED UNIT ACTIVATES
Wed 2021-08-18 12:52:00 EAT 19s left Wed 2021-08-18 12:51:04 EAT 35s ago ocp_csr_approval.timer ocp_csr_approval.service
Wed 2021-08-18 14:07:29 EAT 1h 15min left Wed 2021-08-18 12:43:57 EAT 7min ago dnf-makecache.timer dnf-makecache.service
Thu 2021-08-19 00:00:00 EAT 11h left Wed 2021-08-18 04:17:21 EAT 8h ago unbound-anchor.timer unbound-anchor.service
Thu 2021-08-19 01:16:29 EAT 12h left Wed 2021-08-18 01:16:29 EAT 11h ago systemd-tmpfiles-clean.timer systemd-tmpfiles-clean.service
Mon 2021-08-23 01:09:11 EAT 4 days left Mon 2021-08-16 07:08:52 EAT 2 days ago fstrim.timer fstrim.service
5 timers listed.
Pass --all to see loaded but inactive timers, too.Step 11: Create other OpenShift Users
We have a guide on adding new users to OpenShift Cluster using HTPasswd Provider:
To this point, if you completed the setup steps, should have a running OpenShift Cluster on KVM Virtualization Environment. We will be sharing more guides on Administration and Deployment of various applications in the Cluster. Our aim is to also cover areas around CI/CD and GitOps within OpenShift ecosystem.
More guides on OpenShift:
