openshift/openshift-docs
1
0
Fork 0
mirror of https://github.com/openshift/openshift-docs.git synced 2026-02-05 12:46:18 +01:00
Code Issues Releases Activity
Files
enterprise-4.21
openshift-docs/modules/specifying-host-groups-vsphere.adoc

234 lines
8.4 KiB
Plaintext
Raw Permalink Blame History

// Module included in the following assemblies:
// installing/installing_vsphere/post-install-vsphere-zones-regions-configuration.adoc
:_mod-docs-content-type: PROCEDURE
[id="specifying-host-groups-vsphere_{context}"]
= Specifying multiple host groups for your cluster on vSphere
You can configure the `infrastructures.config.openshift.io` configuration resource to specify multiple host groups for your {product-title} cluster that runs on a {vmw-first} instance. This is necessary if your {vmw-short} instance is in a stretched cluster configuration, with your ESXi hosts and storage distributed across multiple physical data centers. Use this procedure if you did not already configure host groups for your {product-title} cluster at installation, or if you need to update your {product-title} cluster with additional host groups.
:FeatureName: OpenShift zones support for vSphere host groups
include::snippets/technology-preview.adoc[]
.Prerequisites
* ESXi hosts are grouped into host groups, which are linked via VM-host affinity rules to corresponding virtual machine (VM) groups. See the following example `govc` commands for details:
+
[source,terminal]
----
# This example shows the correct configuration for a cluster with two host groups:
# Create host groups:
govc cluster.group.create -name <host_group_1> -host
govc cluster.group.create -name <host_group_2> -host
# Create VM groups:
govc cluster.group.create -name <vm_group_1> -vm
govc cluster.group.create -name <vm_group_2> -vm
# Create VM-host affinity rules:
govc cluster.rule.create -name <rule_1> -enable -vm-host -vm-group <vm_group_1> -host-affine-group <host_group_1>
govc cluster.rule.create -name <rule_2> -enable -vm-host -vm-group <vm_group_2> -host-affine-group <host_group_2>
# Add ESXi hosts to host groups:
govc cluster.group.change -name <host_group_1> <esxi_host_1_ip>
govc cluster.group.change -name <host_group_2> <esxi_host_2_ip>
----
* `openshift-region` and `openshift-zone` tag categories are created on the vCenter server.
* Compute clusters have tags from the `openshift-region` tag category.
* ESXi hosts within host groups have tags from the `openshift-zone` tag category.
* `Host.Inventory.EditCluster` privilege is granted on the {vmw-short} vCenter cluster object.
* `TechPreviewNoUpgrade` feature set is enabled. For more information, "see Enabling features using feature gates".
.Procedure
. Edit the infrastructure settings of your {product-title} cluster.
.. To copy your existing infrastructure settings to a file, run the following command:
+
[source,terminal]
----
$ oc get infrastructures.config.openshift.io cluster -o yaml > <name_of_infrastructure_file>.yaml
----
+
.. Edit your infrastructure file to include a failure domain for each host group in your {vmw-short} cluster. Refer to the following YAML file for an example of this configuration. Ensure you replace any values wrapped in angle brackets (`< >`) with your values:
+
[source,yaml]
----
apiVersion: config.openshift.io/v1
kind: Infrastructure
metadata:
name: cluster
spec:
cloudConfig:
key: config
name: cloud-provider-config
platformSpec:
type: VSphere
vsphere:
apiServerInternalIPs:
- <internal_ip_of_api_server>
failureDomains:
- name: <unique_name_for_failure_domain_1>
region: <cluster_1_region_tag>
server: <vcenter_server_ip_address>
zoneAffinity:
type: HostGroup
hostGroup:
vmGroup: <name_of_vm_group_1>
hostGroup: <name_of_host_group_1>
vmHostRule: <name_of_vm_host_affinity_rule_1>
regionAffinity:
type: ComputeCluster
topology:
computeCluster: /<data_center_1>/host/<cluster_1>
datacenter: <data_center_1>
datastore: /<data_center_1>/datastore/<datastore_1>
networks:
- VM Network
resourcePool: /<data_center_1>/host/<cluster_1>/Resources
template: /<data_center_1>/vm/<vm_template>
zone: <host_group_1_tag>
- name: <unique_name_for_failure_domain_2>
region: <cluster_1_region_tag>
server: <vcenter_server_ip_address>
zoneAffinity:
type: HostGroup
hostGroup:
vmGroup: <name_of_vm_group_2>
hostGroup: <name_of_host_group_2>
vmHostRule: <name_of_vm_host_affinity_rule_2>
regionAffinity:
type: ComputeCluster
topology:
computeCluster: /<data_center_1>/host/<cluster_1>
datacenter: <data_center_1>
datastore: /<data_center_1>/datastore/<datastore_1>
networks:
- VM Network
resourcePool: /<data_center_1>/host/<cluster_1>/Resources
template: /<data_center_1>/vm/<vm_template>
zone: <host_group_2_tag>
# ...
----
+
.. To update your cluster with these changes, run the following command:
+
[source,terminal]
----
$ oc replace -f <name_of_infrastructure_file>.yaml
----
. Update your `ControlPlaneMachineSet` custom resource (CR) with the new failure domains by completing the following steps:
+
.. Edit the `ControlPlaneMachineSet` CR by running the following command:
+
[source,terminal]
----
$ oc edit controlplanemachinesets.machine.openshift.io -n openshift-machine-api cluster
----
+
.. Edit the `failureDomains` parameter as shown in the following example:
+
[source,yaml]
----
spec:
replicas: 3
selector:
matchLabels:
machine.openshift.io/cluster-api-cluster: jdoe3-whb8l
machine.openshift.io/cluster-api-machine-role: master
machine.openshift.io/cluster-api-machine-type: master
state: Active
strategy:
type: RollingUpdate
template:
machineType: machines_v1beta1_machine_openshift_io
machines_v1beta1_machine_openshift_io:
failureDomains:
platform: VSphere
vsphere:
- name: <failure_domain_1_name>
- name: <failure_domain_2_name>
# ...
----
+
.. Verify that your control plane nodes have finished updating before proceeding further. To do this, run the following command:
+
[source,terminal]
----
$ oc get controlplanemachinesets.machine.openshift.io -n openshift-machine-api
----
. Create new `MachineSet` CRs for your failure domains.
+
.. To retrieve the configuration of an existing `MachineSet` CR for use as a template, run the following command:
+
[source,terminal]
----
$ oc get machinesets.machine.openshift.io -n openshift-machine-api <existing_machine_set> -o yaml > machineset-<failure_domain_name>.yaml
----
+
.. Copy the template as needed to create `MachineSet` CR files for each failure domain that you defined in your infrastructure file. Refer to the following example:
+
[source,yaml]
----
apiVersion: machine.openshift.io/v1beta1
kind: MachineSet
metadata:
labels:
machine.openshift.io/cluster-api-cluster: <infrastructure_id>
name: <machineset_name>
namespace: openshift-machine-api
spec:
replicas: 0
selector:
matchLabels:
machine.openshift.io/cluster-api-cluster: <infrastructure_id>
machine.openshift.io/cluster-api-machineset: <machineset_name>
template:
metadata:
labels:
machine.openshift.io/cluster-api-cluster: <infrastructure_id>
machine.openshift.io/cluster-api-machine-role: worker
machine.openshift.io/cluster-api-machine-type: worker
machine.openshift.io/cluster-api-machineset: <machineset_name>
spec:
lifecycleHooks: {}
metadata: {}
providerSpec:
value:
apiVersion: machine.openshift.io/v1beta1
credentialsSecret:
name: vsphere-cloud-credentials
diskGiB: <disk_GiB>
kind: VSphereMachineProviderSpec
memoryMiB: <memory_in_MiB>
metadata:
creationTimestamp: null
network:
devices:
- networkName: VM Network
numCPUs: <number_of_cpus>
numCoresPerSocket: <number_of_cores_per_socket>
snapshot: ""
template: <template_name>
userDataSecret:
name: worker-user-data
workspace:
datacenter: <data_center_1>
datastore: /<data_center_1>/datastore/<datastore_1>
folder: /<data_center_1>/vm/<folder>
resourcePool: /<data_center_1>/host/<cluster_1>/Resources
server: <server_ip_address>
vmGroup: <name_of_vm_group_1>
# ...
----
+
.. For each `MachineSet` CR file, run the following command:
+
[source,terminal]
----
$ oc create -f <name_of_machine_set_file>.yaml
----
View Git Blame Copy Permalink