Red Hat OpenShift Container Platform on HPE SimpliVity

# Deploying RHEL worker nodes

Red Hat Enterprise Linux (RHEL) worker nodes can only be deployed after the OpenShift control plane has been successfully deployed and once the OCP cluster is up and running.

WARNING

If you choose to use RHEL compute machines in your cluster, you take responsibility for all operating system life cycle management and maintenance, including performing system updates, applying patches, and completing all other required tasks.

WARNING

Only RHEL 7.6 is supported in OpenShift Container Platform 4.2. You must not upgrade your compute machines to RHEL 8.

# System requirements for RHEL compute nodes

RHEL compute nodes in your OpenShift Container Platform environment must meet the following minimum hardware specifications and system-level requirements.

  • You must have active OpenShift Container Platform and Red Hat Enterprise Linux Server subscriptions associated with your Red Hat Network account.

  • Production environments must provide compute resources to support your expected application workloads. As an OpenShift Container Platform cluster administrator, you must calculate the resource requirements for your expected workload and add about 10 percent for overhead. For production environments, be sure to allocate enough resources so that your OCP cluster can survive a node failure event without affecting your running applications.

# Running the playbooks

There are a number of steps for deploying RHEL worker nodes:

  • Download Red Hat Ansible playbooks to the folder ~/openshift-ansible folder
  • Run the HPE playbook playbooks/scale.yml in the ~/OpenShift-on-SimpliVity to provision the RHEL VMs
  • Run a Red Hat Ansible playbook to add the RHEL nodes to the OCP cluster.

# Download Red Hat Ansible playbooks

The Red Hat OpenShift ansible playbooks are available at https://github.com/openshift/openshift-ansible. For OpenShift 4.x, this repository only provides playbooks necessary for scaling up or upgrading RHEL hosts in an existing 4.x cluster.

WARNING

The required Red Hat openshift-ansible playbooks change regularly. HPE has no control on these playbooks and therefore recommends that you use a specific version of the playbooks rather than the latest version on the master branch. At the time of writing, the 4.2.2-201910250432 version of the Red Hat openshift-ansible playbooks have been tested and certified with this solution.

Clone the release of the openshift-ansible playbooks that has been tagged openshift-ansible-4.2.2-201910250432 into the home directory of the core user on the Ansible controller node:

$ cd
$ git clone --branch openshift-ansible-4.2.2-201910250432 https://github.com/openshift/openshift-ansible.git

Ensure that the ocp_repo_directory variable in group_vars/all/vars.yml accurately reflects the location of the cloned openshift-ansible playbooks. By default, this parameter is set to: <<local_home>>/openshift-ansible so it should not require modification if the playbooks were cloned in the core user's home directory.

# Running playbooks/scale.yml to provision the RHEL VMs

  • Modify your hosts inventory file to add one or more entries in the [rhel_worker] group. For example, in the extract from the hosts file below, the ocp-worker4 node will be added as a RHEL worker:

    [rhcos_worker]
ocp-worker0   ansible_host=10.15.155.213 
ocp-worker1   ansible_host=10.15.155.214
ocp-worker2   ansible_host=10.15.155.215 
ocp-worker3   ansible_host=10.15.155.216 

[rhel_worker]
ocp-worker4   ansible_host=10.15.155.217

  • Ensure that the value of the template variable in the group variables file group_vars/rhel_worker.yml is set to the name of your RHEL 7.6 template. Alternatively, set the ova_path variable in the same group file to the location of your RHEL 7.6 OVA file. The template variable will take precedence over the ova_path variable if both are present.

    template: ocp-rhel760              # Override the default template
ova_path: ~/kits/ocp-rhel760.ova    # Name of the OVA used to import the template

  • Run the playbooks/scale.yml file

    $ cd ~/OpenShift-on-SimpliVity
$ ansible-playbook -i hosts playbooks/scale.yml --vault-password-file .vault_pass

Note: Running this playbook will generate an inventory file called inventory.scale in the ~/openshift-ansible folder, so that folder must be present before running the playbooks/scale.yml playbook.

# Run the Red Hat Ansible playbook scaleup.yml

Deploy the RHEL worker node(s) by invoking the playbooks/scaleup.yml playbook in the openshift-ansible directory:

$ cd ~/openshift-ansible
$ ansible-playbook -i inventory.scale playbooks/scaleup.yml

Verify that the newly created RHEL worker node ocp-worker4 has joined the cluster:

$ oc get nodes -o wide

NAME          STATUS   ROLES    AGE    VERSION             INTERNAL-IP     EXTERNAL-IP     OS-IMAGE                                                   KERNEL-VERSION               CONTAINER-RUNTIME
ocp-master0   Ready    master   10d    v1.14.6+c7d2111b9   10.15.155.210   10.15.155.210   Red Hat Enterprise Linux CoreOS 42.81.20191107.0 (Ootpa)   4.18.0-147.el8.x86_64        cri-o://1.14.11-0.23.dev.rhaos4.2.gitc41de67.el8
ocp-master1   Ready    master   10d    v1.14.6+c7d2111b9   10.15.155.211   10.15.155.211   Red Hat Enterprise Linux CoreOS 42.81.20191107.0 (Ootpa)   4.18.0-147.el8.x86_64        cri-o://1.14.11-0.23.dev.rhaos4.2.gitc41de67.el8
ocp-master2   Ready    master   10d    v1.14.6+c7d2111b9   10.15.155.212   10.15.155.212   Red Hat Enterprise Linux CoreOS 42.81.20191107.0 (Ootpa)   4.18.0-147.el8.x86_64        cri-o://1.14.11-0.23.dev.rhaos4.2.gitc41de67.el8
ocp-worker0   Ready    worker   10d    v1.14.6+c7d2111b9   10.15.155.213   10.15.155.213   Red Hat Enterprise Linux CoreOS 42.81.20191107.0 (Ootpa)   4.18.0-147.el8.x86_64        cri-o://1.14.11-0.23.dev.rhaos4.2.gitc41de67.el8
ocp-worker1   Ready    worker   10d    v1.14.6+c7d2111b9   10.15.155.214   10.15.155.214   Red Hat Enterprise Linux CoreOS 42.81.20191107.0 (Ootpa)   4.18.0-147.el8.x86_64        cri-o://1.14.11-0.23.dev.rhaos4.2.gitc41de67.el8
ocp-worker2   Ready    worker   8d     v1.14.6+c7d2111b9   10.15.155.215   10.15.155.215   Red Hat Enterprise Linux CoreOS 42.81.20191107.0 (Ootpa)   4.18.0-147.el8.x86_64        cri-o://1.14.11-0.23.dev.rhaos4.2.gitc41de67.el8
ocp-worker3   Ready    worker   8d     v1.14.6+c7d2111b9   10.15.155.216   10.15.155.216   Red Hat Enterprise Linux CoreOS 42.81.20191107.0 (Ootpa)   4.18.0-147.el8.x86_64        cri-o://1.14.11-0.23.dev.rhaos4.2.gitc41de67.el8
ocp-worker4   Ready    worker   7d3h   v1.13.4+509f0153f   10.15.155.217   10.15.155.217   OpenShift Enterprise                                       3.10.0-1062.7.1.el7.x86_64   cri-o://1.14.11-0.17.dev.rhaos4.2.gitc41de67.el7

Note: It can take 10 to 15 minutes to deploy two Red Hat Enterprise Linux worker nodes.

← Introduction to cluster logging →