Configuring Your Inventory File

You are viewing documentation for a release that is no longer supported. The latest supported version of version 3 is [3.11]. For the most recent version 4, see [4]

You are viewing documentation for a release that is no longer supported. The latest supported version of version 3 is [3.11]. For the most recent version 4, see [4]

Customizing Inventory Files for Your Cluster

Ansible inventory files describe the details about the hosts in your cluster, as well as the cluster configuration details for your OKD installation. The OKD installation playbooks read your inventory file to know where and how to install OKD across your set of hosts.

When you install the openshift-ansible RPM package as described in Host Preparation, Ansible dependencies create a file at the default location of /etc/ansible/hosts. However, the file is simply the default Ansible example and has no variables related specifically to OKD configuration. To successfully install OKD, you must replace the default contents of the file with your own desired configuration per your cluster topography and requirements.

The following sections describe commonly used variables to set in your inventory file during cluster installation. Many of the Ansible variables described are optional. Accepting the default values for required variables should suffice for development environments, but for production environments, it is recommended you read through and become familiar with the various options available.

You can review Example Inventory Files for various examples to use as a starting point for your cluster installation.

Configuring Cluster Variables

To assign environment variables during the Ansible install that apply more globally to your OKD cluster overall, indicate the desired variables in the /etc/ansible/hosts file on separate, single lines within the [OSEv3:vars] section. For example:

[OSEv3:vars]
openshift_master_identity_providers=[{'name': 'htpasswd_auth',
'login': 'true', 'challenge': 'true',
'kind': 'HTPasswdPasswordIdentityProvider',}]
openshift_master_default_subdomain=apps.test.example.com

The following tables describe variables for use with the Ansible installer that can be assigned cluster-wide:

openshift_master_admission_plugin_config={“ClusterResourceOverride”:{“configuration”:{“apiVersion”:”v1”,”kind”:”ClusterResourceOverrideConfig”,”memoryRequestToLimitPercent”:”25”,”cpuRequestToLimitPercent”:”25”,”limitCPUToMemoryPercent”:”200”}}}

openshift_master_cluster_public_hostname=openshift-ansible.public.example.com

openshift_docker_additional_registries=example.com:443

Configuring Deployment Type

Various defaults used throughout the playbooks and roles used by the installer are based on the deployment type configuration (usually defined in an Ansible inventory file).

Ensure the openshift_deployment_type parameter in your inventory file’s [OSEv3:vars] section is set to origin to install the OKD variant:

[OSEv3:vars]
openshift_deployment_type=origin

Configuring Host Variables

To assign environment variables to hosts during the Ansible installation, indicate the desired variables in the /etc/ansible/hosts file after the host entry in the [masters] or [nodes] sections. For example:

[masters]
ec2-52-6-179-239.compute-1.amazonaws.com openshift_public_hostname=ose3-master.public.example.com

The following table describes variables for use with the Ansible installer that can be assigned to individual host entries:

OPTIONS=’ —selinux-enabled —log-opt  max-size=1M —log-opt max-file=3 —insecure-registry 172.30.0.0/16 —log-driver=json-file —signature-verification=false’

Defining Node Groups and Host Mappings

Starting in OKD 3.10, node configurations are now bootstrapped from the master. When the node boots and services are started, the node checks if a kubeconfig and other node configuration files exist before joining the cluster. If they do not, the node pulls the configuration from the master, then joins the cluster.

This process replaces administrators having to manually maintain the node configuration uniquely on each node host. Instead, the contents of a node host’s /etc/origin/node/node-config.yaml file are now provided by ConfigMaps sourced from the master.

Node ConfigMaps

The Configmaps for defining the node configurations must be available in the openshift-node project. ConfigMaps are also now the authoritative definition for node labels; the old openshift_node_labels value is effectively ignored.

By default during a cluster installation, the installer creates the following default ConfigMaps:

• node-config-master

• node-config-infra

• node-config-compute

The following ConfigMaps are also created, which label nodes into multiple roles:

• node-config-all-in-one

• node-config-master-infra

The following ConfigMaps are CRI-O variants for each of the existing default node groups:

• node-config-master-crio

• node-config-infra-crio

• node-config-compute-crio

• node-config-all-in-one-crio

• node-config-master-infra-crio

Node Group Definitions

After installing the latest openshift-ansible package, you can view what the default set of node group definitions looks like in YAML format in the ~/openshift-ansible/roles/openshift_facts/defaults/main.yml file:

openshift_node_groups:
  - name: node-config-master (1)
    labels:
      - 'node-role.kubernetes.io/master=true' (2)
    edits: [] (3)
  - name: node-config-infra
    labels:
      - 'node-role.kubernetes.io/infra=true'
    edits: []
  - name: node-config-compute
    labels:
      - 'node-role.kubernetes.io/compute=true'
    edits: []
  - name: node-config-master-infra
    labels:
      - 'node-role.kubernetes.io/infra=true,node-role.kubernetes.io/master=true'
    edits: []
  - name: node-config-all-in-one
    labels:
      - 'node-role.kubernetes.io/infra=true,node-role.kubernetes.io/master=true,node-role.kubernetes.io/compute=true'
    edits: []

If you do not set the openshift_node_groups variable in your inventory file’s [OSEv3:vars] group, the defaults defined above will be used. However, if you want to deviate from these defaults, you must define the entire openshift_node_groups structure (including all desired node groups) in your inventory file.

The openshift_node_groups value is not merged with the defaults, and the YAML definition must first be translated into a Python dictionary. You can then use the edits field to modify any node configuration variables as desired by specifying the key-value pairs.

For example, the following entry in an inventory file defines groups named node-config-master, node-config-infra, and node-config-compute.

openshift_node_groups=[{'name': 'node-config-master', 'labels': ['node-role.kubernetes.io/master=true']}, {'name': 'node-config-infra', 'labels': ['node-role.kubernetes.io/infra=true']}, {'name': 'node-config-compute', 'labels': ['node-role.kubernetes.io/compute=true']}]

You can also define new node group names with other labels, the following entry in an inventory file defines groups named node-config-master, node-config-infra, node-config-compute and node-config-compute-storage.

openshift_node_groups=[{'name': 'node-config-master', 'labels': ['node-role.kubernetes.io/master=true']}, {'name': 'node-config-infra', 'labels': ['node-role.kubernetes.io/infra=true']}, {'name': 'node-config-compute', 'labels': ['node-role.kubernetes.io/compute=true']}, {'name': 'node-config-compute-storage', 'labels': ['node-role.kubernetes.io/compute-storage=true']}]

When you set an entry in the inventory file, you can also edit the ConfigMap for a node group:

• You can use a list, such as modifying the node-config-compute to set kubeletArguments.pods-per-core to 20:

openshift_node_groups=[{'name': 'node-config-master', 'labels': ['node-role.kubernetes.io/master=true']}, {'name': 'node-config-infra', 'labels': ['node-role.kubernetes.io/infra=true']}, {'name': 'node-config-compute', 'labels': ['node-role.kubernetes.io/compute=true'], 'edits': [{ 'key': 'kubeletArguments.pods-per-core','value': ['20']}]}]

• You can use a list to modify multiple key value pairs, such as modifying the node-config-compute group to add two parameters to the kubelet:

openshift_node_groups=[{'name': 'node-config-master', 'labels': ['node-role.kubernetes.io/master=true']}, {'name': 'node-config-infra', 'labels': ['node-role.kubernetes.io/infra=true']}, {'name': 'node-config-compute', 'labels': ['node-role.kubernetes.io/compute=true'], 'edits': [{ 'key': 'kubeletArguments.experimental-allocatable-ignore-eviction','value': ['true']}, {'key': 'kubeletArguments.eviction-hard', 'value': ['memory.available<1Ki']}]}]

• You can use also use a dictionary as value, such as modifying the node-config-compute group to set perFSGroup to 512Mi:

openshift_node_groups=[{'name': 'node-config-master', 'labels': ['node-role.kubernetes.io/master=true']}, {'name': 'node-config-infra', 'labels': ['node-role.kubernetes.io/infra=true']}, {'name': 'node-config-compute', 'labels': ['node-role.kubernetes.io/compute=true'], 'edits': [{ 'key': 'volumeConfig.localQuota','value': {'perFSGroup':'512Mi'}}]}]

Whenever the openshift_node_group.yml playbook is run, the changes defined in the edits field will update the related ConfigMap (node-config-compute in this example), which will ultimately affect the node’s configuration file on the host.

Mapping Hosts to Node Groups

To map which ConfigMap to use for which node host, all hosts defined in the [nodes] group of your inventory must be assigned to a node group using the openshift_node_group_name variable.

The value of openshift_node_group_name is used to select the ConfigMap that configures each node. For example:

[nodes]
master[1:3].example.com openshift_node_group_name='node-config-master'
infra-node1.example.com openshift_node_group_name='node-config-infra'
infra-node2.example.com openshift_node_group_name='node-config-infra'
node1.example.com openshift_node_group_name='node-config-compute'
node2.example.com openshift_node_group_name='node-config-compute'

If other custom ConfigMaps have been defined in openshift_node_groups they can also be used. For exmaple:

[nodes]
master[1:3].example.com openshift_node_group_name='node-config-master'
infra-node1.example.com openshift_node_group_name='node-config-infra'
infra-node2.example.com openshift_node_group_name='node-config-infra'
node1.example.com openshift_node_group_name='node-config-compute'
node2.example.com openshift_node_group_name='node-config-compute'
gluster[1:6].example.com openshift_node_group_name='node-config-compute-storage'

Node Host Labels

Labels can be assigned to node hosts during cluster installation, which are useful for determining the placement of pods onto nodes using the scheduler. While previously node labels could be set using the openshift_node_labels variable, starting in OKD 3.10 you must instead create your own custom node groups if you want to modify the default labels that are assigned to node hosts. See Node Group Definitions for details on modifying the default node groups.

Other than node-role.kubernetes.io/infra=true (hosts using this group are also referred to as dedicated infrastructure nodes and discussed further in Configuring Dedicated Infrastructure Nodes), the actual label names and values are arbitrary and can be assigned however you see fit per your cluster’s requirements.

Pod Schedulability on Masters

Any hosts you designate as masters during the installation process should also be configured as nodes so that the masters are configured as part of the OpenShift SDN. You must do so by adding entries for these hosts to the [nodes] section:

[nodes]
master[1:3].example.com openshift_node_group_name='node-config-master'

If you want to change the schedulability of a host post-installation, see Marking Nodes as Unschedulable or Schedulable.

Pod Schedulability on Nodes

Masters are marked as schedulable nodes by default, so the default node selector is set by default during cluster installations. The default node selector is defined in the master configuration file’s projectConfig.defaultNodeSelector field to determine which node projects will use by default when placing pods. It is set to node-role.kubernetes.io/compute=true unless overridden using the osm_default_node_selector variable.

See Setting the Cluster-wide Default Node Selector for steps on adjusting this setting post-installation if needed.

Configuring Dedicated Infrastructure Nodes

It is recommended for production environments that you maintain dedicated infrastructure nodes where the registry and router pods can run separately from pods used for user applications.

The openshift_router_selector and openshift_registry_selector Ansible settings determine the label selectors used when placing registry and router pods. They are set to node-role.kubernetes.io/infra=true by default:

# default selectors for router and registry services
# openshift_router_selector='node-role.kubernetes.io/infra=true'
# openshift_registry_selector='node-role.kubernetes.io/infra=true'

The registry and router are only able to run on node hosts with the node-role.kubernetes.io/infra=true label, which are then considered dedicated infrastructure nodes. Ensure that at least one node host in your OKD environment has the node-role.kubernetes.io/infra=true label; you can use the default node-config-infra, which sets this label:

[nodes]
infra-node1.example.com openshift_node_group_name='node-config-infra'

If you do not intend to use OKD to manage the registry and router, configure the following Ansible settings:

openshift_hosted_manage_registry=false
openshift_hosted_manage_router=false

If you are using an image registry other than the default registry.access.redhat.com, you must specify the desired registry in the /etc/ansible/hosts file.

As described in Configuring Schedulability on Masters, master hosts are marked schedulable by default. If you label a master host with node-role.kubernetes.io/infra=true and have no other dedicated infrastructure nodes, the master hosts must also be marked as schedulable. Otherwise, the registry and router pods cannot be placed anywhere.

You can use the default node-config-master-infra node group to achieve this:

[nodes]
master.example.com openshift_node_group_name='node-config-master-infra'

Configuring Project Parameters

To configure the default project settings, configure the following variables in the /etc/ansible/hosts file:

Configuring Master API Port

To configure the default ports used by the master API, configure the following variables in the /etc/ansible/hosts file:

For example:

openshift_master_api_port=3443

The web console port setting (openshift_master_console_port) must match the API server port (openshift_master_api_port).

Configuring Cluster Pre-install Checks

Pre-install checks are a set of diagnostic tasks that run as part of the openshift_health_checker Ansible role. They run prior to an Ansible installation of OKD, ensure that required inventory values are set, and identify potential issues on a host that can prevent or interfere with a successful installation.

The following table describes available pre-install checks that will run before every Ansible installation of OKD:

To disable specific pre-install checks, include the variable openshift_disable_check with a comma-delimited list of check names in your inventory file. For example:

openshift_disable_check=memory_availability,disk_availability

Configuring a Registry Location

If you are using an image registry other than the default at registry.access.redhat.com, specify the desired registry within the /etc/ansible/hosts file.

oreg_url=example.com/openshift3/ose-${component}:${version}
openshift_examples_modify_imagestreams=true

For example:

oreg_url=example.com/openshift3/ose-${component}:${version}
oreg_auth_user=${user_name}
oreg_auth_password=${password}
openshift_examples_modify_imagestreams=true

Configuring a Registry Route

To allow users to push and pull images to the internal Docker registry from outside of the OKD cluster, configure the registry route in the /etc/ansible/hosts file. By default, the registry route is docker-registry-default.router.default.svc.cluster.local.

For example:

openshift_hosted_registry_routehost=<path>
openshift_hosted_registry_routetermination=reencrypt
openshift_hosted_registry_routecertificates= "{'certfile': '<path>/org-cert.pem', 'keyfile': '<path>/org-privkey.pem', 'cafile': '<path>/org-chain.pem'}"

Configuring the Registry Console

If you are using a Cockpit registry console image other than the default or require a specific version of the console, specify the desired registry within the /etc/ansible/hosts file:

openshift_cockpit_deployer_prefix=<registry_name>/<namespace>/
openshift_cockpit_deployer_version=<cockpit_image_tag>

For example: If your image is at registry.example.com/openshift3/registry-console and you require version 3.10.1, enter:

openshift_cockpit_deployer_prefix='registry.example.com/openshift3/'
openshift_cockpit_deployer_version='3.10.1'

Configuring Router Sharding

Router sharding support is enabled by supplying the correct data to the inventory. The variable openshift_hosted_routers holds the data, which is in the form of a list. If no data is passed, then a default router is created. There are multiple combinations of router sharding. The following example supports routers on separate nodes:

openshift_hosted_routers=[{'name': 'router1', 'certificate': {'certfile': '/path/to/certificate/abc.crt',
'keyfile': '/path/to/certificate/abc.key', 'cafile':
'/path/to/certificate/ca.crt'}, 'replicas': 1, 'serviceaccount': 'router',
'namespace': 'default', 'stats_port': 1936, 'edits': [], 'images':
'openshift3/ose-${component}:${version}', 'selector': 'type=router1', 'ports':
['80:80', '443:443']},
{'name': 'router2', 'certificate': {'certfile': '/path/to/certificate/xyz.crt',
'keyfile': '/path/to/certificate/xyz.key', 'cafile':
'/path/to/certificate/ca.crt'}, 'replicas': 1, 'serviceaccount': 'router',
'namespace': 'default', 'stats_port': 1936, 'edits': [{'action': 'append',
'key': 'spec.template.spec.containers[0].env', 'value': {'name': 'ROUTE_LABELS',
'value': 'route=external'}}], 'images':
'openshift3/ose-${component}:${version}', 'selector': 'type=router2', 'ports':
['80:80', '443:443']}]

Configuring GlusterFS Persistent Storage

GlusterFS can be configured to provide persistent storage and dynamic provisioning for OKD. It can be used both containerized within OKD (Containerized GlusterFS) and non-containerized on its own nodes (External GlusterFS).

Additional information and examples, including the ones below, can be found at Persistent Storage Using GlusterFS.

Configuring Containerized GlusterFS

1. In your inventory file, include the following variables in the [OSEv3:vars] section, and adjust them as required for your configuration:

   [OSEv3:vars]
   ...
   openshift_storage_glusterfs_namespace=app-storage
   openshift_storage_glusterfs_storageclass=true
   openshift_storage_glusterfs_storageclass_default=false
   openshift_storage_glusterfs_block_deploy=true
   openshift_storage_glusterfs_block_host_vol_size=100
   openshift_storage_glusterfs_block_storageclass=true
   openshift_storage_glusterfs_block_storageclass_default=false

2. Add glusterfs in the [OSEv3:children] section to enable the [glusterfs] group:

   [OSEv3:children]
   masters
   nodes
   glusterfs

3. Add a [glusterfs] section with entries for each storage node that will host the GlusterFS storage. For each node, set glusterfs_devices to a list of raw block devices that will be completely managed as part of a GlusterFS cluster. There must be at least one device listed. Each device must be bare, with no partitions or LVM PVs. Specifying the variable takes the form:

   <hostname_or_ip> glusterfs_devices='[ "</path/to/device1/>", "</path/to/device2>", ... ]'

   For example:

   [glusterfs]
   node11.example.com glusterfs_devices='[ "/dev/xvdc", "/dev/xvdd" ]'
   node12.example.com glusterfs_devices='[ "/dev/xvdc", "/dev/xvdd" ]'
   node13.example.com glusterfs_devices='[ "/dev/xvdc", "/dev/xvdd" ]'

4. Add the hosts listed under [glusterfs] to the [nodes] group:

   [nodes]
   ...
   node11.example.com openshift_node_group_name="node-config-compute"
   node12.example.com openshift_node_group_name="node-config-compute"
   node13.example.com openshift_node_group_name="node-config-compute"

Configuring External GlusterFS

1. In your inventory file, include the following variables in the [OSEv3:vars] section, and adjust them as required for your configuration:

   [OSEv3:vars]
   ...
   openshift_storage_glusterfs_namespace=app-storage
   openshift_storage_glusterfs_storageclass=true
   openshift_storage_glusterfs_storageclass_default=false
   openshift_storage_glusterfs_block_deploy=true
   openshift_storage_glusterfs_block_host_vol_size=100
   openshift_storage_glusterfs_block_storageclass=true
   openshift_storage_glusterfs_block_storageclass_default=false
   openshift_storage_glusterfs_is_native=false
   openshift_storage_glusterfs_heketi_is_native=true
   openshift_storage_glusterfs_heketi_executor=ssh
   openshift_storage_glusterfs_heketi_ssh_port=22
   openshift_storage_glusterfs_heketi_ssh_user=root
   openshift_storage_glusterfs_heketi_ssh_sudo=false
   openshift_storage_glusterfs_heketi_ssh_keyfile="/root/.ssh/id_rsa"

2. Add glusterfs in the [OSEv3:children] section to enable the [glusterfs] group:

   [OSEv3:children]
   masters
   nodes
   glusterfs

3. Add a [glusterfs] section with entries for each storage node that will host the GlusterFS storage. For each node, set glusterfs_devices to a list of raw block devices that will be completely managed as part of a GlusterFS cluster. There must be at least one device listed. Each device must be bare, with no partitions or LVM PVs. Also, set glusterfs_ip to the IP address of the node. Specifying the variable takes the form:

   <hostname_or_ip> glusterfs_ip=<ip_address> glusterfs_devices='[ "</path/to/device1/>", "</path/to/device2>", ... ]'

   For example:

   [glusterfs]
   gluster1.example.com glusterfs_ip=192.168.10.11 glusterfs_devices='[ "/dev/xvdc", "/dev/xvdd" ]'
   gluster2.example.com glusterfs_ip=192.168.10.12 glusterfs_devices='[ "/dev/xvdc", "/dev/xvdd" ]'
   gluster3.example.com glusterfs_ip=192.168.10.13 glusterfs_devices='[ "/dev/xvdc", "/dev/xvdd" ]'

Configuring an OpenShift Container Registry

An integrated OpenShift Container Registry can be deployed using the installer.

Configuring Registry Storage

If no registry storage options are used, the default OpenShift Container Registry is ephemeral and all data will be lost when the pod no longer exists.

There are several options for enabling registry storage when using the advanced installer:

Option A: NFS Host Group

When the following variables are set, an NFS volume is created during cluster installation with the path <nfs_directory>/<volume_name> on the host within the [nfs] host group. For example, the volume path using these options would be /exports/registry:

[OSEv3:vars]
openshift_hosted_registry_storage_kind=nfs
openshift_hosted_registry_storage_access_modes=['ReadWriteMany']
openshift_hosted_registry_storage_nfs_directory=/exports
openshift_hosted_registry_storage_nfs_options='*(rw,root_squash)'
openshift_hosted_registry_storage_volume_name=registry
openshift_hosted_registry_storage_volume_size=10Gi

Option B: External NFS Host

To use an external NFS volume, one must already exist with a path of <nfs_directory>/<volume_name> on the storage host. The remote volume path using the following options would be nfs.example.com:/exports/registry.

[OSEv3:vars]
openshift_hosted_registry_storage_kind=nfs
openshift_hosted_registry_storage_access_modes=['ReadWriteMany']
openshift_hosted_registry_storage_host=nfs.example.com
openshift_hosted_registry_storage_nfs_directory=/exports
openshift_hosted_registry_storage_volume_name=registry
openshift_hosted_registry_storage_volume_size=10Gi

Upgrading or Installing OKD with NFS

Option C: OpenStack Platform

An OpenStack storage configuration must already exist.

[OSEv3:vars]
openshift_hosted_registry_storage_kind=openstack
openshift_hosted_registry_storage_access_modes=['ReadWriteOnce']
openshift_hosted_registry_storage_openstack_filesystem=ext4
openshift_hosted_registry_storage_openstack_volumeID=3a650b4f-c8c5-4e0a-8ca5-eaee11f16c57
openshift_hosted_registry_storage_volume_size=10Gi

Option D: AWS or Another S3 Storage Solution

The simple storage solution (S3) bucket must already exist.

[OSEv3:vars]
#openshift_hosted_registry_storage_kind=object
#openshift_hosted_registry_storage_provider=s3
#openshift_hosted_registry_storage_s3_accesskey=access_key_id
#openshift_hosted_registry_storage_s3_secretkey=secret_access_key
#openshift_hosted_registry_storage_s3_bucket=bucket_name
#openshift_hosted_registry_storage_s3_region=bucket_region
#openshift_hosted_registry_storage_s3_chunksize=26214400
#openshift_hosted_registry_storage_s3_rootdirectory=/registry
#openshift_hosted_registry_pullthrough=true
#openshift_hosted_registry_acceptschema2=true
#openshift_hosted_registry_enforcequota=true

If you are using a different S3 service, such as Minio or ExoScale, also add the region endpoint parameter:

openshift_hosted_registry_storage_s3_regionendpoint=https://myendpoint.example.com/

Option E: Containerized GlusterFS

Similar to configuring Containerized GlusterFS, GlusterFS can be configured to provide storage for an OpenShift Container Registry during the initial installation of the cluster to offer redundant and reliable storage for the registry.

1. In your inventory file, set the following variable under [OSEv3:vars] section, and adjust them as required for your configuration:

   [OSEv3:vars]
   ...
   openshift_hosted_registry_storage_kind=glusterfs (1)
   openshift_hosted_registry_storage_volume_size=5Gi
   openshift_hosted_registry_selector='node-role.kubernetes.io/infra=true'

   1	Running the integrated OpenShift Container Registry, on infrastructure nodes is recommended. Infrastructure node are nodes dedicated to running applications deployed by administrators to provide services for the OKD cluster.

2. Add glusterfs_registry in the [OSEv3:children] section to enable the [glusterfs_registry] group:

   [OSEv3:children]
   masters
   nodes
   glusterfs_registry

3. Add a [glusterfs_registry] section with entries for each storage node that will host the GlusterFS storage. For each node, set glusterfs_devices to a list of raw block devices that will be completely managed as part of a GlusterFS cluster. There must be at least one device listed. Each device must be bare, with no partitions or LVM PVs. Specifying the variable takes the form:

   <hostname_or_ip> glusterfs_devices='[ "</path/to/device1/>", "</path/to/device2>", ... ]'

   For example:

   [glusterfs_registry]
   node11.example.com glusterfs_devices='[ "/dev/xvdc", "/dev/xvdd" ]'
   node12.example.com glusterfs_devices='[ "/dev/xvdc", "/dev/xvdd" ]'
   node13.example.com glusterfs_devices='[ "/dev/xvdc", "/dev/xvdd" ]'

4. Add the hosts listed under [glusterfs_registry] to the [nodes] group:

   [nodes]
   ...
   node11.example.com openshift_node_group_name="node-config-infra"
   node12.example.com openshift_node_group_name="node-config-infra"
   node13.example.com openshift_node_group_name="node-config-infra"

Option F: Google Cloud Storage (GCS) bucket on Google Compute Engine (GCE)

A GCS bucket must already exist.

[OSEv3:vars]
openshift_hosted_registry_storage_provider=gcs
openshift_hosted_registry_storage_gcs_bucket=bucket01
openshift_hosted_registry_storage_gcs_keyfile=test.key
openshift_hosted_registry_storage_gcs_rootdirectory=/registry

Option G: vSphere Volume with vSphere Cloud Provider (VCP)

The vSphere Cloud Provider must be configured with a datastore accessible by the OKD nodes.

When using vSphere volume for the registry, you must set the storage access mode to ReadWriteOnce and the replica count to 1:

[OSEv3:vars]
openshift_hosted_registry_storage_kind=vsphere
openshift_hosted_registry_storage_access_modes=['ReadWriteOnce']
openshift_hosted_registry_storage_annotations=['volume.beta.kubernetes.io/storage-provisioner: kubernetes.io/vsphere-volume']
openshift_hosted_registry_replicas=1

Configuring Global Proxy Options

If your hosts require use of a HTTP or HTTPS proxy in order to connect to external hosts, there are many components that must be configured to use the proxy, including masters, Docker, and builds. Node services only connect to the master API requiring no external access and therefore do not need to be configured to use a proxy.

In order to simplify this configuration, the following Ansible variables can be specified at a cluster or host level to apply these settings uniformly across your environment.

Configuring the Firewall

OKD uses iptables as the default firewall, but you can configure your cluster to use firewalld during the install process.

Because iptables is the default firewall, OKD is designed to have it configured automatically. However, iptables rules can break OKD if not configured correctly. The advantages of firewalld include allowing multiple objects to safely share the firewall rules.

To use firewalld as the firewall for an OKD installation, add the os_firewall_use_firewalld variable to the list of configuration variables in the Ansible host file at install:

[OSEv3:vars]
os_firewall_use_firewalld=True (1)

Configuring Session Options

Session options in the OAuth configuration are configurable in the inventory file. By default, Ansible populates a sessionSecretsFile with generated authentication and encryption secrets so that sessions generated by one master can be decoded by the others. The default location is /etc/origin/master/session-secrets.yaml, and this file will only be re-created if deleted on all masters.

You can set the session name and maximum number of seconds with openshift_master_session_name and openshift_master_session_max_seconds:

openshift_master_session_name=ssn
openshift_master_session_max_seconds=3600

If provided, openshift_master_session_auth_secrets and openshift_master_encryption_secrets must be equal length.

For openshift_master_session_auth_secrets, used to authenticate sessions using HMAC, it is recommended to use secrets with 32 or 64 bytes:

openshift_master_session_auth_secrets=['DONT+USE+THIS+SECRET+b4NV+pmZNSO']

For openshift_master_encryption_secrets, used to encrypt sessions, secrets must be 16, 24, or 32 characters long, to select AES-128, AES-192, or AES-256:

openshift_master_session_encryption_secrets=['DONT+USE+THIS+SECRET+b4NV+pmZNSO']

Configuring Custom Certificates

Custom serving certificates for the public host names of the OKD API and web console can be deployed during cluster installation and are configurable in the inventory file.

Certificate and key file paths can be configured using the openshift_master_named_certificates cluster variable:

openshift_master_named_certificates=[{"certfile": "/path/to/custom1.crt", "keyfile": "/path/to/custom1.key", "cafile": "/path/to/custom-ca1.crt"}]

File paths must be local to the system where Ansible will be run. Certificates are copied to master hosts and are deployed within the /etc/origin/master/named_certificates/ directory.

Ansible detects a certificate’s Common Name and Subject Alternative Names. Detected names can be overridden by providing the "names" key when setting openshift_master_named_certificates:

openshift_master_named_certificates=[{"certfile": "/path/to/custom1.crt", "keyfile": "/path/to/custom1.key", "names": ["public-master-host.com"], "cafile": "/path/to/custom-ca1.crt"}]

Certificates configured using openshift_master_named_certificates are cached on masters, meaning that each additional Ansible run with a different set of certificates results in all previously deployed certificates remaining in place on master hosts and within the master configuration file.

If you would like openshift_master_named_certificates to be overwritten with the provided value (or no value), specify the openshift_master_overwrite_named_certificates cluster variable:

openshift_master_overwrite_named_certificates=true

For a more complete example, consider the following cluster variables in an inventory file:

openshift_master_cluster_method=native
openshift_master_cluster_hostname=lb-internal.openshift.com
openshift_master_cluster_public_hostname=custom.openshift.com

To overwrite the certificates on a subsequent Ansible run, you could set the following:

openshift_master_named_certificates=[{"certfile": "/root/STAR.openshift.com.crt", "keyfile": "/root/STAR.openshift.com.key", "names": ["custom.openshift.com"]}]
openshift_master_overwrite_named_certificates=true

Configuring Certificate Validity

By default, the certificates used to govern the etcd, master, and kubelet expire after two to five years. The validity (length in days until they expire) for the auto-generated registry, CA, node, and master certificates can be configured during installation using the following variables (default values shown):

[OSEv3:vars]
openshift_hosted_registry_cert_expire_days=730
openshift_ca_cert_expire_days=1825
openshift_node_cert_expire_days=730
openshift_master_cert_expire_days=730
etcd_ca_default_days=1825

These values are also used when redeploying certificates via Ansible post-installation.

Configuring Cluster Metrics

Cluster metrics are not set to automatically deploy. Set the following to enable cluster metrics during cluster installation:

[OSEv3:vars]
openshift_metrics_install_metrics=true

The metrics public URL can be set during cluster installation using the openshift_metrics_hawkular_hostname Ansible variable, which defaults to:

https://hawkular-metrics.{{openshift_master_default_subdomain}}/hawkular/metrics

If you alter this variable, ensure the host name is accessible via your router.

openshift_metrics_hawkular_hostname=hawkular-metrics.{{openshift_master_default_subdomain}}

Configuring Metrics Storage

The openshift_metrics_cassandra_storage_type variable must be set in order to use persistent storage for metrics. If openshift_metrics_cassandra_storage_type is not set, then cluster metrics data is stored in an emptyDir volume, which will be deleted when the Cassandra pod terminates.

There are three options for enabling cluster metrics storage during cluster installation:

Option A: Dynamic

If your OKD environment supports dynamic volume provisioning for your cloud provider, use the following variable:

[OSEv3:vars]
openshift_metrics_cassandra_storage_type=dynamic

If there are multiple default dynamically provisioned volume types, such as gluster-storage and glusterfs-storage-block, you can specify the provisioned volume type by variable. Use the following variables:

[OSEv3:vars]
openshift_metrics_cassandra_storage_type=pv
openshift_metrics_cassandra_pvc_storage_class_name=glusterfs-storage-block

Check Volume Configuration for more information on using DynamicProvisioningEnabled to enable or disable dynamic provisioning.

Option B: NFS Host Group

When the following variables are set, an NFS volume is created during cluster installation with path <nfs_directory>/<volume_name> on the host within the [nfs] host group. For example, the volume path using these options would be /exports/metrics:

[OSEv3:vars]
openshift_metrics_storage_kind=nfs
openshift_metrics_storage_access_modes=['ReadWriteOnce']
openshift_metrics_storage_nfs_directory=/exports
openshift_metrics_storage_nfs_options='*(rw,root_squash)'
openshift_metrics_storage_volume_name=metrics
openshift_metrics_storage_volume_size=10Gi

Option C: External NFS Host

To use an external NFS volume, one must already exist with a path of <nfs_directory>/<volume_name> on the storage host.

[OSEv3:vars]
openshift_metrics_storage_kind=nfs
openshift_metrics_storage_access_modes=['ReadWriteOnce']
openshift_metrics_storage_host=nfs.example.com
openshift_metrics_storage_nfs_directory=/exports
openshift_metrics_storage_volume_name=metrics
openshift_metrics_storage_volume_size=10Gi

The remote volume path using the following options would be nfs.example.com:/exports/metrics.

Upgrading or Installing OKD with NFS

The use of NFS for the core OKD components is not recommended, as NFS (and the NFS Protocol) does not provide the proper consistency needed for the applications that make up the OKD infrastructure.

As a result, the installer and update playbooks require an option to enable the use of NFS with core infrastructure components.

# Enable unsupported configurations, things that will yield a partially
# functioning cluster but would not be supported for production use
#openshift_enable_unsupported_configurations=false

If you see the following messages when upgrading or installing your cluster, then an additional step is required.

TASK [Run variable sanity checks] **********************************************
fatal: [host.example.com]: FAILED! => {"failed": true, "msg": "last_checked_host: host.example.com, last_checked_var: openshift_hosted_registry_storage_kind;nfs is an unsupported type for openshift_hosted_registry_storage_kind. openshift_enable_unsupported_configurations=True mustbe specified to continue with this configuration."}

In your Ansible inventory file, specify the following parameter:

[OSEv3:vars]
openshift_enable_unsupported_configurations=True

Configuring Cluster Logging

Cluster logging is not set to automatically deploy by default. Set the following to enable cluster logging during cluster installation:

[OSEv3:vars]
openshift_logging_install_logging=true

Configuring Logging Storage

The openshift_logging_es_pvc_dynamic variable must be set in order to use persistent storage for logging. If openshift_logging_es_pvc_dynamic is not set, then cluster logging data is stored in an emptyDir volume, which will be deleted when the Elasticsearch pod terminates.

There are three options for enabling cluster logging storage during cluster installation:

Option A: Dynamic

If your OKD environment has dynamic volume provisioning, it could be configured either via the cloud provider or by an independent storage provider. For instance, the cloud provider could have a StorageClass with provisioner kubernetes.io/gce-pd on GCE, and an independent storage provider such as GlusterFS could have a StorageClass with provisioner kubernetes.io/glusterfs. In either case, use the following variable:

[OSEv3:vars]
openshift_logging_es_pvc_dynamic=true

For additional information on dynamic provisioning, see Dynamic provisioning and creating storage classes.

If there are multiple default dynamically provisioned volume types, such as gluster-storage and glusterfs-storage-block, you can specify the provisioned volume type by variable. Use the following variables:

[OSEv3:vars]
openshift_logging_elasticsearch_storage_type=pvc
openshift_logging_es_pvc_storage_class_name=glusterfs-storage-block

Check Volume Configuration for more information on using DynamicProvisioningEnabled to enable or disable dynamic provisioning.

Option B: NFS Host Group

When the following variables are set, an NFS volume is created during cluster installation with path <nfs_directory>/<volume_name> on the host within the [nfs] host group. For example, the volume path using these options would be /exports/logging:

[OSEv3:vars]
openshift_logging_storage_kind=nfs
openshift_logging_storage_access_modes=['ReadWriteOnce']
openshift_logging_storage_nfs_directory=/exports
openshift_logging_storage_nfs_options='*(rw,root_squash)'
openshift_logging_storage_volume_name=logging
openshift_logging_storage_volume_size=10Gi

Option C: External NFS Host

To use an external NFS volume, one must already exist with a path of <nfs_directory>/<volume_name> on the storage host.

[OSEv3:vars]
openshift_logging_storage_kind=nfs
openshift_logging_storage_access_modes=['ReadWriteOnce']
openshift_logging_storage_host=nfs.example.com
openshift_logging_storage_nfs_directory=/exports
openshift_logging_storage_volume_name=logging
openshift_logging_storage_volume_size=10Gi

The remote volume path using the following options would be nfs.example.com:/exports/logging.

Upgrading or Installing OKD with NFS

The use of NFS for the core OKD components is not recommended, as NFS (and the NFS Protocol) does not provide the proper consistency needed for the applications that make up the OKD infrastructure.

As a result, the installer and update playbooks require an option to enable the use of NFS with core infrastructure components.

# Enable unsupported configurations, things that will yield a partially
# functioning cluster but would not be supported for production use
#openshift_enable_unsupported_configurations=false

If you see the following messages when upgrading or installing your cluster, then an additional step is required.

TASK [Run variable sanity checks] **********************************************
fatal: [host.example.com]: FAILED! => {"failed": true, "msg": "last_checked_host: host.example.com, last_checked_var: openshift_hosted_registry_storage_kind;nfs is an unsupported type for openshift_hosted_registry_storage_kind. openshift_enable_unsupported_configurations=True mustbe specified to continue with this configuration."}

In your Ansible inventory file, specify the following parameter:

[OSEv3:vars]
openshift_enable_unsupported_configurations=True

Customizing Service Catalog Options

The service catalog is enabled by default during installation. Enabling the service broker allows you to register service brokers with the catalog. When the service catalog is enabled, the OpenShift Ansible broker and template service broker are both installed as well; see Configuring the OpenShift Ansible Broker and Configuring the Template Service Broker for more information. If you disable the service catalog, the OpenShift Ansible broker and template service broker are not installed.

To disable automatic deployment of the service catalog, set the following cluster variable in your inventory file:

openshift_enable_service_catalog=false

If you use your own registry, you must add:

• openshift_service_catalog_image_prefix: When pulling the service catalog image, force the use of a specific prefix (for example, registry). You must provide the full registry name up to the image name.

• openshift_service_catalog_image_version: When pulling the service catalog image, force the use of a specific image version.

For example:

openshift_service_catalog_image="docker-registry.default.example.com/openshift/ose-service-catalog:${version}"
openshift_service_catalog_image_prefix="docker-registry-default.example.com/openshift/ose-"
openshift_service_catalog_image_version="v3.9.30"
template_service_broker_selector={"role":"infra"}

Configuring the OpenShift Ansible Broker

The OpenShift Ansible broker (OAB) is enabled by default during installation.

If you do not want to install the OAB, set the ansible_service_broker_install parameter value to false in the inventory file:

ansible_service_broker_install=false

Configuring Persistent Storage for the OpenShift Ansible Broker

The OAB deploys its own etcd instance separate from the etcd used by the rest of the OKD cluster. The OAB’s etcd instance requires separate storage using persistent volumes (PVs) to function. If no PV is available, etcd will wait until the PV can be satisfied. The OAB application will enter a CrashLoop state until its etcd instance is available.

Some Ansible playbook bundles (APBs) also require a PV for their own usage in order to deploy. For example, each of the database APBs have two plans: the Development plan uses ephemeral storage and does not require a PV, while the Production plan is persisted and does require a PV.

To configure persistent storage for the OAB:

1. In your inventory file, add nfs to the [OSEv3:children] section to enable the [nfs] group:

   [OSEv3:children]
   masters
   nodes
   nfs

2. Add a [nfs] group section and add the host name for the system that will be the NFS host:

   [nfs]
   master1.example.com

3. Add the following in the [OSEv3:vars] section:

   openshift_hosted_etcd_storage_kind=nfs
   openshift_hosted_etcd_storage_nfs_options="*(rw,root_squash,sync,no_wdelay)"
   openshift_hosted_etcd_storage_nfs_directory=/opt/osev3-etcd (1)
   openshift_hosted_etcd_storage_volume_name=etcd-vol2 (1)
   openshift_hosted_etcd_storage_access_modes=["ReadWriteOnce"]
   openshift_hosted_etcd_storage_volume_size=1G
   openshift_hosted_etcd_storage_labels={'storage': 'etcd'}
   ansible_service_broker_registry_url=registry.access.redhat.com
   ansible_service_broker_registry_user=<user_name> (2)
   ansible_service_broker_registry_password=<password> (2)
   ansible_service_broker_registry_organization=<organization> (2)

   1	An NFS volume will be created with path <nfs_directory>/<volume_name> on the host within the [nfs] group. For example, the volume path using these options would be /opt/osev3-etcd/etcd-vol2.
   2	Only required if ansible_service_broker_registry_url is set to a registry that requires authentication for pulling APBs.

   These settings create a persistent volume that is attached to the OAB’s etcd instance during cluster installation.

Configuring the OpenShift Ansible Broker for Local APB Development

In order to do APB development with the OpenShift Container Registry in conjunction with the OAB, a whitelist of images the OAB can access must be defined. If a whitelist is not defined, the broker will ignore APBs and users will not see any APBs available.

By default, the whitelist is empty so that a user cannot add APB images to the broker without a cluster administrator configuring the broker. To whitelist all images that end in -apb:

1. In your inventory file, add the following to the [OSEv3:vars] section:

   ansible_service_broker_local_registry_whitelist=['.*-apb$']

Configuring the Template Service Broker

The template service broker (TSB) is enabled by default during installation.

If you do not want to install the TSB, set the template_service_broker_install parameter value to false:

template_service_broker_install=false

To configure the TSB, one or more projects must be defined as the broker’s source namespace(s) for loading templates and image streams into the service catalog. Set the desired projects by modifying the following in your inventory file’s [OSEv3:vars] section:

openshift_template_service_broker_namespaces=['openshift','myproject']

By default, the TSB will use the nodeselector {"node-role.kubernetes.io/infra":"true"} for deploying its pods. You can modify this by setting the desired nodeselector in your inventory file’s [OSEv3:vars] section:

template_service_broker_selector={"node-role.kubernetes.io/infra":"true"}

Configuring Web Console Customization

The following Ansible variables set master configuration options for customizing the web console. See Customizing the Web Console for more details on these customization options.