Configuring for Red Hat Virtualization | Configuring Clusters

Configuring Red Hat Virtualization objects
Configuring OKD for Red Hat Virtualization

You can configure OKD to use Red Hat Virtualization.

Configuring Red Hat Virtualization objects

To integrate OKD with Red Hat Virtualization, take the following actions as part of your host preparation.

To provide high availability in case of the loss of one hypervisor host, add each class of instance to a negative affinity group. See VM Affinity.

To ensure that the OKD environment meets the minimum hardware requirements, create templates for virtual machines that use the following resources:

Master nodes

Minimum 2 CPU Cores
16 GB Memory
Minimum 10 GB root disk
Minimum 15 GB container image registry disk
30 GB local volume disk
Minimum 25 GB etcd disk

Infrastructure nodes

Minimum 2 CPU Cores
16 GB Memory
Minimum 10 GB root disk
Minimum 15 GB container image registry disk
30 GB local volume disk
Minimum 25 GB Gluster registry disk

Application nodes

2 CPU Cores
8 GB Memory
Minimum 10 GB root disk
Minimum 15 GB container image registry disk
30 GB local volume disk

Load balancer node

1 CPU Core
4 GB Memory
10 GB root disk

Create master, infrastructure, and application nodes as well as a load balancer node. Use the templates that you created.
Create DNS entries for the routers. Provide entries for all infrastructure instances and configure a round-robin strategy so that the router can pass traffic to applications.
Create a DNS entry for the OKD web console. Specify the IP address of the load balancer node.
To use Red Hat Virtualization, you must provide external storage, for example GlusterFS, for persistent storage of registry images and for application storage.

Configuring OKD for Red Hat Virtualization

You configure OKD for Red Hat Virtualization by modifying the Ansible inventory file before you install the cluster.

Modify the Ansible inventory file, located at /etc/ansible/hosts by default, to use the following YAML sections:

[OSEv3:children]
nodes
masters
etcd
glusterfs_registry
lb

[OSEv3:vars]
# General variables
ansible_ssh_user=root
openshift_deployment_type=openshift-enterprise
openshift_release='3.10'
openshift_master_cluster_method=native
debug_level=2
openshift_debug_level="{{ debug_level }}"
openshift_node_debug_level="{{ node_debug_level | default(debug_level, true) }}"
openshift_enable_service_catalog=False

app_dns_prefix=apps
public_hosted_zone=example.com
load_balancer_hostname=lb.{{public_hosted_zone}}
openshift_master_cluster_hostname="{{ load_balancer_hostname }}"
openshift_master_cluster_public_hostname="{{ load_balancer_hostname }}"
openshift_master_default_subdomain="{{ app_dns_prefix }}.{{ public_hosted_zone }}"

# Pod Networking
os_sdn_network_plugin_name=redhat/openshift-ovs-networkpolicy

# Registry
openshift_hosted_registry_storage_kind=glusterfs

# Authentication (example here creates one user, myuser with password changeme)
openshift_master_identity_providers="[{'name': 'htpasswd_auth', 'login': 'true', 'challenge': 'True', 'kind': 'HTPasswdPasswordIdentityProvider', 'filename': '/etc/origin/master/htpasswd'}]"
openshift_master_htpasswd_users={'myuser': '$apr1$zAhyA9Ko$rBxBOwAwwtRuuaw8OtCwH0'}

# Docker and extra file system setup
container_runtime_docker_storage_setup_device=/dev/vdb
container_runtime_docker_storage_type=overlay2
openshift_docker_use_system_container=False
openshift_node_local_quota_per_fsgroup=512Mi (1)
openshift_use_system_containers=False

[masters]
master0.example.com
master1.example.com
master2.example.com

[etcd]
master0.example.com
master1.example.com
master2.example.com

[infras]
infra0.example.com
infra1.example.com
infra2.example.com

[glusterfs_registry]
infra0.example.com glusterfs_devices="['/dev/vdd']"
infra1.example.com glusterfs_devices="['/dev/vdd']"
infra2.example.com glusterfs_devices="['/dev/vdd']"

[lb]
lb.example.com

[nodes]
master0.example.com openshift_node_group_name=node-config-master
master1.example.com openshift_node_group_name=node-config-master
master2.example.com openshift_node_group_name=node-config-master
infra0.example.com openshift_node_group_name=node-config-infra
infra1.example.com openshift_node_group_name=node-config-infra
infra2.example.com openshift_node_group_name=node-config-infra
app0.example.com openshift_node_group_name=node-config-compute
app1.example.com openshift_node_group_name=node-config-compute
app2.example.com openshift_node_group_name=node-config-compute

1	If you use the `openshift_node_local_quota_per_fsgroup` parameter, you must specify the partition or LVM to use for the directory of `/var/lib/origin/openshift.local.volumes`. The partition must be mounted with the option of `gquota` in fstab.

This inventory file uses the following nodes and disks:

One load balancer instance
Three master instances
- Extra disks attached: 15 GB for the container image registry, 30 GB for local volume storage, and 25 GB for etcd
Three infrastructure instances
- Extra disks attached: 15 GB for Docker, 30 GB for local volume storage, and, because this cluster uses GlusterFS for persistent storage, 25 GB for a GlusterFS registry
One or more application instance
- Extra disks attached: 15 GB for Docker, 30 GB for local volume storage

Continue to install the cluster following the Installing OKD steps. During that process, make any changes to your inventory file that your cluster needs.