This is a cache of https://docs.okd.io/latest/installing/installing_azure/ipi/installing-azure-private.html. It is a snapshot of the page at 2024-11-23T18:05:37.456+0000.
Installing a private cluster - Installing on Azure | Installing | OKD 4
  1. Documentation
×

Installing a private cluster on Azure

Private clusters

You can deploy a private OKD cluster that does not expose external endpoints. Private clusters are accessible from only an internal network and are not visible to the internet.

By default, OKD is provisioned to use publicly-accessible DNS and endpoints. A private cluster sets the DNS, Ingress Controller, and API server to private when you deploy your cluster. This means that the cluster resources are only accessible from your internal network and are not visible to the internet.

If the cluster has any public subnets, load balancer services created by administrators might be publicly accessible. To ensure cluster security, verify that these services are explicitly annotated as private.

To deploy a private cluster, you must:

  • Use existing networking that meets your requirements. Your cluster resources might be shared between other clusters on the network.

  • Deploy from a machine that has access to:

    • The API services for the cloud to which you provision.

    • The hosts on the network that you provision.

    • The internet to obtain installation media.

You can use any machine that meets these access requirements and follows your company’s guidelines. For example, this machine can be a bastion host on your cloud network or a machine that has access to the network through a VPN.

Private clusters in Azure

To create a private cluster on Microsoft Azure, you must provide an existing private VNet and subnets to host the cluster. The installation program must also be able to resolve the DNS records that the cluster requires. The installation program configures the Ingress Operator and API server for only internal traffic.

Depending how your network connects to the private VNET, you might need to use a DNS forwarder to resolve the cluster’s private DNS records. The cluster’s machines use 168.63.129.16 internally for DNS resolution. For more information, see What is Azure Private DNS? and What is IP address 168.63.129.16? in the Azure documentation.

The cluster still requires access to internet to access the Azure APIs.

The following items are not required or created when you install a private cluster:

  • A BaseDomainResourceGroup, since the cluster does not create public records

  • Public IP addresses

  • Public DNS records

  • Public endpoints

    The cluster is configured so that the Operators do not create public records for the cluster and all cluster machines are placed in the private subnets that you specify.

Limitations

Private clusters on Azure are subject to only the limitations that are associated with the use of an existing VNet.

User-defined outbound routing

In OKD, you can choose your own outbound routing for a cluster to connect to the internet. This allows you to skip the creation of public IP addresses and the public load balancer.

You can configure user-defined routing by modifying parameters in the install-config.yaml file before installing your cluster. A pre-existing VNet is required to use outbound routing when installing a cluster; the installation program is not responsible for configuring this.

When configuring a cluster to use user-defined routing, the installation program does not create the following resources:

  • Outbound rules for access to the internet.

  • Public IPs for the public load balancer.

  • Kubernetes Service object to add the cluster machines to the public load balancer for outbound requests.

You must ensure the following items are available before setting user-defined routing:

  • Egress to the internet is possible to pull container images, unless using an OpenShift image registry mirror.

  • The cluster can access Azure APIs.

  • Various allowlist endpoints are configured. You can reference these endpoints in the Configuring your firewall section.

There are several pre-existing networking setups that are supported for internet access using user-defined routing.

Private cluster with network address translation

You can use Azure VNET network address translation (NAT) to provide outbound internet access for the subnets in your cluster. You can reference Create a NAT gateway using Azure CLI in the Azure documentation for configuration instructions.

When using a VNet setup with Azure NAT and user-defined routing configured, you can create a private cluster with no public endpoints.

Private cluster with Azure Firewall

You can use Azure Firewall to provide outbound routing for the VNet used to install the cluster. You can learn more about providing user-defined routing with Azure Firewall in the Azure documentation.

When using a VNet setup with Azure Firewall and user-defined routing configured, you can create a private cluster with no public endpoints.

Private cluster with a proxy configuration

You can use a proxy with user-defined routing to allow egress to the internet. You must ensure that cluster Operators do not access Azure APIs using a proxy; Operators must have access to Azure APIs outside of the proxy.

When using the default route table for subnets, with 0.0.0.0/0 populated automatically by Azure, all Azure API requests are routed over Azure’s internal network even though the IP addresses are public. As long as the Network Security Group rules allow egress to Azure API endpoints, proxies with user-defined routing configured allow you to create private clusters with no public endpoints.

Private cluster with no internet access

You can install a private network that restricts all access to the internet, except the Azure API. This is accomplished by mirroring the release image registry locally. Your cluster must have access to the following:

  • An OpenShift image registry mirror that allows for pulling container images

  • Access to Azure APIs

With these requirements available, you can use user-defined routing to create private clusters with no public endpoints.

About reusing a VNet for your OKD cluster

In OKD 4, you can deploy a cluster into an existing Azure Virtual Network (VNet) in Microsoft Azure. If you do, you must also use existing subnets within the VNet and routing rules.

By deploying OKD into an existing Azure VNet, you might be able to avoid service limit constraints in new accounts or more easily abide by the operational constraints that your company’s guidelines set. This is a good option to use if you cannot obtain the infrastructure creation permissions that are required to create the VNet.

Requirements for using your VNet

When you deploy a cluster by using an existing VNet, you must perform additional network configuration before you install the cluster. In installer-provisioned infrastructure clusters, the installer usually creates the following components, but it does not create them when you install into an existing VNet:

  • Subnets

  • Route tables

  • VNets

  • Network Security Groups

The installation program requires that you use the cloud-provided DNS server. Using a custom DNS server is not supported and causes the installation to fail.

If you use a custom VNet, you must correctly configure it and its subnets for the installation program and the cluster to use. The installation program cannot subdivide network ranges for the cluster to use, set route tables for the subnets, or set VNet options like DHCP, so you must do so before you install the cluster.

The cluster must be able to access the resource group that contains the existing VNet and subnets. While all of the resources that the cluster creates are placed in a separate resource group that it creates, some network resources are used from a separate group. Some cluster Operators must be able to access resources in both resource groups. For example, the Machine API controller attaches NICS for the virtual machines that it creates to subnets from the networking resource group.

Your VNet must meet the following characteristics:

  • The VNet’s CIDR block must contain the Networking.MachineCIDR range, which is the IP address pool for cluster machines.

  • The VNet and its subnets must belong to the same resource group, and the subnets must be configured to use Azure-assigned DHCP IP addresses instead of static IP addresses.

You must provide two subnets within your VNet, one for the control plane machines and one for the compute machines. Because Azure distributes machines in different availability zones within the region that you specify, your cluster will have high availability by default.

By default, if you specify availability zones in the install-config.yaml file, the installation program distributes the control plane machines and the compute machines across these availability zones within a region. To ensure high availability for your cluster, select a region with at least three availability zones. If your region contains fewer than three availability zones, the installation program places more than one control plane machine in the available zones.

To ensure that the subnets that you provide are suitable, the installation program confirms the following data:

  • All the specified subnets exist.

  • There are two private subnets, one for the control plane machines and one for the compute machines.

  • The subnet CIDRs belong to the machine CIDR that you specified. Machines are not provisioned in availability zones that you do not provide private subnets for.

If you destroy a cluster that uses an existing VNet, the VNet is not deleted.

Network security group requirements

The network security groups for the subnets that host the compute and control plane machines require specific access to ensure that the cluster communication is correct. You must create rules to allow access to the required cluster communication ports.

The network security group rules must be in place before you install the cluster. If you attempt to install a cluster without the required access, the installation program cannot reach the Azure APIs, and installation fails.
Table 1. Required ports
Port Description Control plane Compute

80

Allows HTTP traffic

x

443

Allows HTTPS traffic

x

6443

Allows communication to the control plane machines

x

22623

Allows internal communication to the machine config server for provisioning machines

x

  1. If you are using Azure Firewall to restrict the internet access, then you can configure Azure Firewall to allow the Azure APIs. A network security group rule is not needed. For more information, see "Configuring your firewall" in "Additional resources".

Currently, there is no supported way to block or restrict the machine config server endpoint. The machine config server must be exposed to the network so that newly-provisioned machines, which have no existing configuration or state, are able to fetch their configuration. In this model, the root of trust is the certificate signing requests (CSR) endpoint, which is where the kubelet sends its certificate signing request for approval to join the cluster. Because of this, machine configs should not be used to distribute sensitive information, such as secrets and certificates.

To ensure that the machine config server endpoints, ports 22623 and 22624, are secured in bare metal scenarios, customers must configure proper network policies.

Because cluster components do not modify the user-provided network security groups, which the Kubernetes controllers update, a pseudo-network security group is created for the Kubernetes controller to modify without impacting the rest of the environment.

Table 2. Ports used for all-machine to all-machine communications
Protocol Port Description

ICMP

N/A

Network reachability tests

TCP

1936

Metrics

9000-9999

Host level services, including the node exporter on ports 9100-9101 and the Cluster Version Operator on port 9099.

10250-10259

The default ports that Kubernetes reserves

UDP

4789

VXLAN

6081

Geneve

9000-9999

Host level services, including the node exporter on ports 9100-9101.

500

IPsec IKE packets

4500

IPsec NAT-T packets

123

Network Time Protocol (ntp) on UDP port 123

If you configure an external ntp time server, you must open UDP port 123.

TCP/UDP

30000-32767

Kubernetes node port

ESP

N/A

IPsec Encapsulating Security Payload (ESP)
Table 3. Ports used for control plane machine to control plane machine communications
Protocol Port Description

TCP

2379-2380

etcd server and peer ports

Division of permissions

Starting with OKD 4.3, you do not need all of the permissions that are required for an installation program-provisioned infrastructure cluster to deploy a cluster. This change mimics the division of permissions that you might have at your company: some individuals can create different resources in your clouds than others. For example, you might be able to create application-specific items, like instances, storage, and load balancers, but not networking-related components such as VNets, subnet, or ingress rules.

The Azure credentials that you use when you create your cluster do not need the networking permissions that are required to make VNets and core networking components within the VNet, such as subnets, routing tables, internet gateways, NAT, and VPN. You still need permission to make the application resources that the machines within the cluster require, such as load balancers, security groups, storage accounts, and nodes.

Isolation between clusters

Because the cluster is unable to modify network security groups in an existing subnet, there is no way to isolate clusters from each other on the VNet.

Additional resources

Manually creating the installation configuration file

Installing the cluster requires that you manually create the installation configuration file.

Prerequisites

  • You have an SSH public key on your local machine to provide to the installation program. The key will be used for SSH authentication onto your cluster nodes for debugging and disaster recovery.

  • You have obtained the OKD installation program and the pull secret for your cluster.

Procedure

  1. Create an installation directory to store your required installation assets in:

    $ mkdir <installation_directory>

    You must create a directory. Some installation assets, like bootstrap X.509 certificates have short expiration intervals, so you must not reuse an installation directory. If you want to reuse individual files from another cluster installation, you can copy them into your directory. However, the file names for the installation assets might change between releases. Use caution when copying installation files from an earlier OKD version.

  2. Customize the sample install-config.yaml file template that is provided and save it in the <installation_directory>.

    You must name this configuration file install-config.yaml.

  3. Back up the install-config.yaml file so that you can use it to install multiple clusters.

    The install-config.yaml file is consumed during the next step of the installation process. You must back it up now.
Additional resources

Minimum resource requirements for cluster installation

Each cluster machine must meet the following minimum requirements:

Table 4. Minimum resource requirements
Machine Operating System vCPU [1] Virtual RAM Storage Input/Output Per Second (IOPS)[2]

Bootstrap

FCOS

4

16 GB

100 GB

300

Control plane

FCOS

4

16 GB

100 GB

300

Compute

FCOS

2

8 GB

100 GB

300

  1. One vCPU is equivalent to one physical core when simultaneous multithreading (SMT), or Hyper-Threading, is not enabled. When enabled, use the following formula to calculate the corresponding ratio: (threads per core × cores) × sockets = vCPUs.

  2. OKD and Kubernetes are sensitive to disk performance, and faster storage is recommended, particularly for etcd on the control plane nodes which require a 10 ms p99 fsync duration. Note that on many cloud platforms, storage size and IOPS scale together, so you might need to over-allocate storage volume to obtain sufficient performance.

  3. As with all user-provisioned installations, if you choose to use Fedora 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. Use of Fedora 7 compute machines is deprecated and has been removed in OKD 4.10 and later.

As of OKD version 4.13, RHCOS is based on RHEL version 9.2, which updates the micro-architecture requirements. The following list contains the minimum instruction set architectures (ISA) that each architecture requires:

  • x86-64 architecture requires x86-64-v2 ISA

  • ARM64 architecture requires ARMv8.0-A ISA

  • IBM Power architecture requires Power 9 ISA

  • s390x architecture requires z14 ISA

For more information, see RHEL Architectures.

You are required to use Azure virtual machines that have the premiumIO parameter set to true.

If an instance type for your platform meets the minimum requirements for cluster machines, it is supported to use in OKD.

Additional resources

Tested instance types for Azure

The following Microsoft Azure instance types have been tested with OKD.

Machine types based on 64-bit x86 architecture

  • standardBasv2Family

  • standardBSFamily

  • standardBsv2Family

  • standardDADSv5Family

  • standardDASv4Family

  • standardDASv5Family

  • standardDCACCV5Family

  • standardDCADCCV5Family

  • standardDCADSv5Family

  • standardDCASv5Family

  • standardDCSv3Family

  • standardDCSv2Family

  • standardDDCSv3Family

  • standardDDSv4Family

  • standardDDSv5Family

  • standardDLDSv5Family

  • standardDLSv5Family

  • standardDSFamily

  • standardDSv2Family

  • standardDSv2PromoFamily

  • standardDSv3Family

  • standardDSv4Family

  • standardDSv5Family

  • standardEADSv5Family

  • standardEASv4Family

  • standardEASv5Family

  • standardEBDSv5Family

  • standardEBSv5Family

  • standardECACCV5Family

  • standardECADCCV5Family

  • standardECADSv5Family

  • standardECASv5Family

  • standardEDSv4Family

  • standardEDSv5Family

  • standardEIADSv5Family

  • standardEIASv4Family

  • standardEIASv5Family

  • standardEIBDSv5Family

  • standardEIBSv5Family

  • standardEIDSv5Family

  • standardEISv3Family

  • standardEISv5Family

  • standardESv3Family

  • standardESv4Family

  • standardESv5Family

  • standardFXMDVSFamily

  • standardFSFamily

  • standardFSv2Family

  • standardGSFamily

  • standardHBrsv2Family

  • standardHBSFamily

  • standardHBv4Family

  • standardHCSFamily

  • standardHXFamily

  • standardLASv3Family

  • standardLSFamily

  • standardLSv2Family

  • standardLSv3Family

  • standardMDSHighMemoryv3Family

  • standardMDSMediumMemoryv2Family

  • standardMDSMediumMemoryv3Family

  • standardMIDSHighMemoryv3Family

  • standardMIDSMediumMemoryv2Family

  • standardMISHighMemoryv3Family

  • standardMISMediumMemoryv2Family

  • standardMSFamily

  • standardMSHighMemoryv3Family

  • standardMSMediumMemoryv2Family

  • standardMSMediumMemoryv3Family

  • StandardNCADSA100v4Family

  • Standard NCASv3_T4 Family

  • standardNCSv3Family

  • standardNDSv2Family

  • StandardNGADSV620v1Family

  • standardNPSFamily

  • StandardNVADSA10v5Family

  • standardNVSv3Family

  • standardXEISv4Family

Tested instance types for Azure on 64-bit ARM infrastructures

The following Microsoft Azure ARM64 instance types have been tested with OKD.

Machine types based on 64-bit ARM architecture

  • standardBpsv2Family

  • standardDPSv5Family

  • standardDPDSv5Family

  • standardDPLDSv5Family

  • standardDPLSv5Family

  • standardEPSv5Family

  • standardEPDSv5Family

Enabling trusted launch for Azure VMs

You can enable two trusted launch features when installing your cluster on Azure: secure boot and virtualized Trusted Platform Modules.

For more information about the sizes of virtual machines that support the trusted launch features, see Virtual machine sizes.

Trusted launch is a Technology Preview feature only. Technology Preview features are not supported with Red Hat production service level agreements (SLAs) and might not be functionally complete. Red Hat does not recommend using them in production. These features provide early access to upcoming product features, enabling customers to test functionality and provide feedback during the development process.

For more information about the support scope of Red Hat Technology Preview features, see Technology Preview Features Support Scope.
Prerequisites

  • You have created an install-config.yaml file.

Procedure

  • Edit the install-config.yaml file before deploying your cluster:

    • Enable trusted launch only on control plane by adding the following stanza:

      controlPlane:
  platform:
    azure:
      settings:
        securityType: TrustedLaunch
        trustedLaunch:
          uefiSettings:
            secureBoot: Enabled
            virtualizedTrustedPlatformModule: Enabled

    • Enable trusted launch only on compute node by adding the following stanza:

      compute:
  platform:
    azure:
      settings:
        securityType: TrustedLaunch
        trustedLaunch:
          uefiSettings:
            secureBoot: Enabled
            virtualizedTrustedPlatformModule: Enabled

    • Enable trusted launch on all nodes by adding the following stanza:

      platform:
  azure:
    settings:
      securityType: TrustedLaunch
      trustedLaunch:
        uefiSettings:
          secureBoot: Enabled
          virtualizedTrustedPlatformModule: Enabled

Enabling confidential VMs

You can enable confidential VMs when installing your cluster. You can enable confidential VMs for compute nodes, control plane nodes, or all nodes.

Using confidential VMs is a Technology Preview feature only. Technology Preview features are not supported with Red Hat production service level agreements (SLAs) and might not be functionally complete. Red Hat does not recommend using them in production. These features provide early access to upcoming product features, enabling customers to test functionality and provide feedback during the development process.

For more information about the support scope of Red Hat Technology Preview features, see Technology Preview Features Support Scope.

You can use confidential VMs with the following VM sizes:

  • DCasv5-series

  • DCadsv5-series

  • ECasv5-series

  • ECadsv5-series

Confidential VMs are currently not supported on 64-bit ARM architectures.
Prerequisites

  • You have created an install-config.yaml file.

Procedure

  • Edit the install-config.yaml file before deploying your cluster:

    • Enable confidential VMs only on control plane by adding the following stanza:

      controlPlane:
  platform:
    azure:
      settings:
        securityType: ConfidentialVM
        confidentialVM:
          uefiSettings:
            secureBoot: Enabled
            virtualizedTrustedPlatformModule: Enabled
      osDisk:
        securityProfile:
          securityEncryptionType: VMGuestStateOnly

    • Enable confidential VMs only on compute nodes by adding the following stanza:

      compute:
  platform:
    azure:
      settings:
        securityType: ConfidentialVM
        confidentialVM:
          uefiSettings:
            secureBoot: Enabled
            virtualizedTrustedPlatformModule: Enabled
      osDisk:
        securityProfile:
          securityEncryptionType: VMGuestStateOnly

    • Enable confidential VMs on all nodes by adding the following stanza:

      platform:
  azure:
    settings:
      securityType: ConfidentialVM
      confidentialVM:
        uefiSettings:
          secureBoot: Enabled
          virtualizedTrustedPlatformModule: Enabled
    osDisk:
      securityProfile:
        securityEncryptionType: VMGuestStateOnly

Sample customized install-config.yaml file for Azure

You can customize the install-config.yaml file to specify more details about your OKD cluster’s platform or modify the values of the required parameters.

This sample YAML file is provided for reference only. You must obtain your install-config.yaml file by using the installation program and modify it.

 
apiVersion: v1
baseDomain: example.com (1)
controlPlane: (2)
  hyperthreading: Enabled  (3) (4)
  name: master
  platform:
    azure:
      encryptionAtHost: true
      ultraSSDCapability: Enabled
      osDisk:
        diskSizeGB: 1024 (5)
        diskType: Premium_LRS
        diskEncryptionSet:
          resourceGroup: disk_encryption_set_resource_group
          name: disk_encryption_set_name
          subscriptionId: secondary_subscription_id
      osImage:
        publisher: example_publisher_name
        offer: example_image_offer
        sku: example_offer_sku
        version: example_image_version
      type: Standard_D8s_v3
  replicas: 3
compute: (2)
- hyperthreading: Enabled  (3) (4)
  name: worker
  platform:
    azure:
      ultraSSDCapability: Enabled
      type: Standard_D2s_v3
      encryptionAtHost: true
      osDisk:
        diskSizeGB: 512 (5)
        diskType: Standard_LRS
        diskEncryptionSet:
          resourceGroup: disk_encryption_set_resource_group
          name: disk_encryption_set_name
          subscriptionId: secondary_subscription_id
      osImage:
        publisher: example_publisher_name
        offer: example_image_offer
        sku: example_offer_sku
        version: example_image_version
      zones: (6)
      - "1"
      - "2"
      - "3"
  replicas: 5
metadata:
  name: test-cluster (1)
networking:
  clusterNetwork:
  - cidr: 10.128.0.0/14
    hostPrefix: 23
  machineNetwork:
  - cidr: 10.0.0.0/16
  networkType: OVNKubernetes (7)
  serviceNetwork:
  - 172.30.0.0/16
platform:
  azure:
    defaultMachinePlatform:
      osImage: (8)
        publisher: example_publisher_name
        offer: example_image_offer
        sku: example_offer_sku
        version: example_image_version
      ultraSSDCapability: Enabled
    baseDomainResourceGroupName: resource_group (9)
    region: centralus (1)
    resourceGroupName: existing_resource_group (10)
    networkResourceGroupName: vnet_resource_group (11)
    virtualNetwork: vnet (12)
    controlPlaneSubnet: control_plane_subnet (13)
    computeSubnet: compute_subnet (14)
    outboundType: UserDefinedRouting (15)
    cloudName: AzurePublicCloud
pullSecret: '{"auths": ...}' (1)
sshKey: ssh-ed25519 AAAA... (16)
publish: Internal (17)
1 Required. The installation program prompts you for this value.
2 If you do not provide these parameters and values, the installation program provides the default value.
3 The controlPlane section is a single mapping, but the compute section is a sequence of mappings. To meet the requirements of the different data structures, the first line of the compute section must begin with a hyphen, -, and the first line of the controlPlane section must not. Only one control plane pool is used.
4 Whether to enable or disable simultaneous multithreading, or hyperthreading. By default, simultaneous multithreading is enabled to increase the performance of your machines' cores. You can disable it by setting the parameter value to Disabled. If you disable simultaneous multithreading in some cluster machines, you must disable it in all cluster machines.

If you disable simultaneous multithreading, ensure that your capacity planning accounts for the dramatically decreased machine performance. Use larger virtual machine types, such as Standard_D8s_v3, for your machines if you disable simultaneous multithreading.
5 You can specify the size of the disk to use in GB. Minimum recommendation for control plane nodes is 1024 GB.
6 Specify a list of zones to deploy your machines to. For high availability, specify at least two zones.
7 The cluster network plugin to install. The default value OVNKubernetes is the only supported value.
8 Optional: A custom Fedora CoreOS (FCOS) image that should be used to boot control plane and compute machines. The publisher, offer, sku, and version parameters under platform.azure.defaultMachinePlatform.osImage apply to both control plane and compute machines. If the parameters under controlPlane.platform.azure.osImage or compute.platform.azure.osImage are set, they override the platform.azure.defaultMachinePlatform.osImage parameters.
9 Specify the name of the resource group that contains the DNS zone for your base domain.
10 Specify the name of an already existing resource group to install your cluster to. If undefined, a new resource group is created for the cluster.
11 If you use an existing VNet, specify the name of the resource group that contains it.
12 If you use an existing VNet, specify its name.
13 If you use an existing VNet, specify the name of the subnet to host the control plane machines.
14 If you use an existing VNet, specify the name of the subnet to host the compute machines.
15 You can customize your own outbound routing. Configuring user-defined routing prevents exposing external endpoints in your cluster. User-defined routing for egress requires deploying your cluster to an existing VNet.
16 You can optionally provide the sshKey value that you use to access the machines in your cluster.

For production OKD clusters on which you want to perform installation debugging or disaster recovery, specify an SSH key that your ssh-agent process uses.
17 How to publish the user-facing endpoints of your cluster. Set publish to Internal to deploy a private cluster, which cannot be accessed from the internet. The default value is External.

Configuring the cluster-wide proxy during installation

Production environments can deny direct access to the internet and instead have an HTTP or HTTPS proxy available. You can configure a new OKD cluster to use a proxy by configuring the proxy settings in the install-config.yaml file.

Prerequisites

  • You have an existing install-config.yaml file.

  • You reviewed the sites that your cluster requires access to and determined whether any of them need to bypass the proxy. By default, all cluster egress traffic is proxied, including calls to hosting cloud provider APIs. You added sites to the Proxy object’s spec.noProxy field to bypass the proxy if necessary.

    The Proxy object status.noProxy field is populated with the values of the networking.machineNetwork[].cidr, networking.clusterNetwork[].cidr, and networking.serviceNetwork[] fields from your installation configuration.

    For installations on Amazon Web Services (AWS), Google Cloud Platform (GCP), Microsoft Azure, and OpenStack, the Proxy object status.noProxy field is also populated with the instance metadata endpoint (169.254.169.254).
Procedure

  1. Edit your install-config.yaml file and add the proxy settings. For example:

    apiVersion: v1
baseDomain: my.domain.com
proxy:
  httpProxy: http://<username>:<pswd>@<ip>:<port> (1)
  httpsProxy: https://<username>:<pswd>@<ip>:<port> (2)
  noProxy: example.com (3)
additionalTrustBundle: | (4)
    -----BEGIN CERTIFICATE-----
    <MY_TRUSTED_CA_CERT>
    -----END CERTIFICATE-----
additionalTrustBundlePolicy: <policy_to_add_additionalTrustBundle> (5)
    1 A proxy URL to use for creating HTTP connections outside the cluster. The URL scheme must be http.
    2 A proxy URL to use for creating HTTPS connections outside the cluster.
    3 A comma-separated list of destination domain names, IP addresses, or other network CIDRs to exclude from proxying. Preface a domain with . to match subdomains only. For example, .y.com matches x.y.com, but not y.com. Use * to bypass the proxy for all destinations.
    4 If provided, the installation program generates a config map that is named user-ca-bundle in the openshift-config namespace to hold the additional CA certificates. If you provide additionalTrustBundle and at least one proxy setting, the Proxy object is configured to reference the user-ca-bundle config map in the trustedCA field. The Cluster Network Operator then creates a trusted-ca-bundle config map that merges the contents specified for the trustedCA parameter with the FCOS trust bundle. The additionalTrustBundle field is required unless the proxy’s identity certificate is signed by an authority from the FCOS trust bundle.
    5 Optional: The policy to determine the configuration of the Proxy object to reference the user-ca-bundle config map in the trustedCA field. The allowed values are Proxyonly and Always. Use Proxyonly to reference the user-ca-bundle config map only when http/https proxy is configured. Use Always to always reference the user-ca-bundle config map. The default value is Proxyonly.

    The installation program does not support the proxy readinessEndpoints field.

    If the installer times out, restart and then complete the deployment by using the wait-for command of the installer. For example:

    $ ./openshift-install wait-for install-complete --log-level debug

  2. Save the file and reference it when installing OKD.

The installation program creates a cluster-wide proxy that is named cluster that uses the proxy settings in the provided install-config.yaml file. If no proxy settings are provided, a cluster Proxy object is still created, but it will have a nil spec.

Only the Proxy object named cluster is supported, and no additional proxies can be created.
Additional resources

Alternatives to storing administrator-level secrets in the kube-system project

By default, administrator secrets are stored in the kube-system project. If you configured the credentialsMode parameter in the install-config.yaml file to Manual, you must use one of the following alternatives:

Manually creating long-term credentials

The Cloud Credential Operator (CCO) can be put into manual mode prior to installation in environments where the cloud identity and access management (IAM) APIs are not reachable, or the administrator prefers not to store an administrator-level credential secret in the cluster kube-system namespace.

Procedure

  1. If you did not set the credentialsMode parameter in the install-config.yaml configuration file to Manual, modify the value as shown:

    Sample configuration file snippet
    apiVersion: v1
baseDomain: example.com
credentialsMode: Manual
# ...

  2. If you have not previously created installation manifest files, do so by running the following command:

    $ openshift-install create manifests --dir <installation_directory>

    where <installation_directory> is the directory in which the installation program creates files.

  3. Set a $RELEASE_IMAGE variable with the release image from your installation file by running the following command:

    $ RELEASE_IMAGE=$(./openshift-install version | awk '/release image/ {print $3}')

  4. Extract the list of CredentialsRequest custom resources (CRs) from the OKD release image by running the following command:

    $ oc adm release extract \
  --from=$RELEASE_IMAGE \
  --credentials-requests \
  --included \(1)
  --install-config=<path_to_directory_with_installation_configuration>/install-config.yaml \(2)
  --to=<path_to_directory_for_credentials_requests> (3)
    1 The --included parameter includes only the manifests that your specific cluster configuration requires.
    2 Specify the location of the install-config.yaml file.
    3 Specify the path to the directory where you want to store the CredentialsRequest objects. If the specified directory does not exist, this command creates it.

    This command creates a YAML file for each CredentialsRequest object.

    Sample CredentialsRequest object
    apiVersion: cloudcredential.openshift.io/v1
kind: CredentialsRequest
metadata:
  name: <component_credentials_request>
  namespace: openshift-cloud-credential-operator
  ...
spec:
  providerSpec:
    apiVersion: cloudcredential.openshift.io/v1
    kind: AzureProviderSpec
    roleBindings:
    - role: Contributor
  ...

  5. Create YAML files for secrets in the openshift-install manifests directory that you generated previously. The secrets must be stored using the namespace and secret name defined in the spec.secretRef for each CredentialsRequest object.

    Sample CredentialsRequest object with secrets
    apiVersion: cloudcredential.openshift.io/v1
kind: CredentialsRequest
metadata:
  name: <component_credentials_request>
  namespace: openshift-cloud-credential-operator
  ...
spec:
  providerSpec:
    apiVersion: cloudcredential.openshift.io/v1
    kind: AzureProviderSpec
    roleBindings:
    - role: Contributor
      ...
  secretRef:
    name: <component_secret>
    namespace: <component_namespace>
  ...
    Sample Secret object
    apiVersion: v1
kind: Secret
metadata:
  name: <component_secret>
  namespace: <component_namespace>
data:
  azure_subscription_id: <base64_encoded_azure_subscription_id>
  azure_client_id: <base64_encoded_azure_client_id>
  azure_client_secret: <base64_encoded_azure_client_secret>
  azure_tenant_id: <base64_encoded_azure_tenant_id>
  azure_resource_prefix: <base64_encoded_azure_resource_prefix>
  azure_resourcegroup: <base64_encoded_azure_resourcegroup>
  azure_region: <base64_encoded_azure_region>

Before upgrading a cluster that uses manually maintained credentials, you must ensure that the CCO is in an upgradeable state.

Configuring an Azure cluster to use short-term credentials

To install a cluster that uses Microsoft Entra Workload ID, you must configure the Cloud Credential Operator utility and create the required Azure resources for your cluster.

Configuring the Cloud Credential Operator utility

To create and manage cloud credentials from outside of the cluster when the Cloud Credential Operator (CCO) is operating in manual mode, extract and prepare the CCO utility (ccoctl) binary.

The ccoctl utility is a Linux binary that must run in a Linux environment.
Prerequisites

  • You have access to an OKD account with cluster administrator access.

  • You have installed the OpenShift CLI (oc).

  • You have created a global Microsoft Azure account for the ccoctl utility to use with the following permissions:

    Required Azure permissions

    • Microsoft.Resources/subscriptions/resourceGroups/read

    • Microsoft.Resources/subscriptions/resourceGroups/write

    • Microsoft.Resources/subscriptions/resourceGroups/delete

    • Microsoft.Authorization/roleAssignments/read

    • Microsoft.Authorization/roleAssignments/delete

    • Microsoft.Authorization/roleAssignments/write

    • Microsoft.Authorization/roleDefinitions/read

    • Microsoft.Authorization/roleDefinitions/write

    • Microsoft.Authorization/roleDefinitions/delete

    • Microsoft.Storage/storageAccounts/listkeys/action

    • Microsoft.Storage/storageAccounts/delete

    • Microsoft.Storage/storageAccounts/read

    • Microsoft.Storage/storageAccounts/write

    • Microsoft.Storage/storageAccounts/blobServices/containers/write

    • Microsoft.Storage/storageAccounts/blobServices/containers/delete

    • Microsoft.Storage/storageAccounts/blobServices/containers/read

    • Microsoft.ManagedIdentity/userAssignedIdentities/delete

    • Microsoft.ManagedIdentity/userAssignedIdentities/read

    • Microsoft.ManagedIdentity/userAssignedIdentities/write

    • Microsoft.ManagedIdentity/userAssignedIdentities/federatedIdentityCredentials/read

    • Microsoft.ManagedIdentity/userAssignedIdentities/federatedIdentityCredentials/write

    • Microsoft.ManagedIdentity/userAssignedIdentities/federatedIdentityCredentials/delete

    • Microsoft.Storage/register/action

    • Microsoft.ManagedIdentity/register/action

Procedure

  1. Set a variable for the OKD release image by running the following command:

    $ RELEASE_IMAGE=$(./openshift-install version | awk '/release image/ {print $3}')

  2. Obtain the CCO container image from the OKD release image by running the following command:

    $ CCO_IMAGE=$(oc adm release info --image-for='cloud-credential-operator' $RELEASE_IMAGE -a ~/.pull-secret)

    Ensure that the architecture of the $RELEASE_IMAGE matches the architecture of the environment in which you will use the ccoctl tool.

  3. Extract the ccoctl binary from the CCO container image within the OKD release image by running the following command:

    $ oc image extract $CCO_IMAGE \
  --file="/usr/bin/ccoctl.<rhel_version>" \(1)
  -a ~/.pull-secret
    1 For <rhel_version>, specify the value that corresponds to the version of Fedora that the host uses. If no value is specified, ccoctl.rhel8 is used by default. The following values are valid:

    • rhel8: Specify this value for hosts that use Fedora 8.

    • rhel9: Specify this value for hosts that use Fedora 9.

  4. Change the permissions to make ccoctl executable by running the following command:

    $ chmod 775 ccoctl.<rhel_version>
Verification

  • To verify that ccoctl is ready to use, display the help file. Use a relative file name when you run the command, for example:

    $ ./ccoctl.rhel9
    Example output
    OpenShift credentials provisioning tool

Usage:
  ccoctl [command]

Available Commands:
  aws          Manage credentials objects for AWS cloud
  azure        Manage credentials objects for Azure
  gcp          Manage credentials objects for Google cloud
  help         Help about any command
  ibmcloud     Manage credentials objects for {ibm-cloud-title}
  nutanix      Manage credentials objects for Nutanix

Flags:
  -h, --help   help for ccoctl

Use "ccoctl [command] --help" for more information about a command.

Creating Azure resources with the Cloud Credential Operator utility

You can use the ccoctl azure create-all command to automate the creation of Azure resources.

By default, ccoctl creates objects in the directory in which the commands are run. To create the objects in a different directory, use the --output-dir flag. This procedure uses <path_to_ccoctl_output_dir> to refer to this directory.
Prerequisites

You must have:

  • Extracted and prepared the ccoctl binary.

  • Access to your Microsoft Azure account by using the Azure CLI.

Procedure

  1. Set a $RELEASE_IMAGE variable with the release image from your installation file by running the following command:

    $ RELEASE_IMAGE=$(./openshift-install version | awk '/release image/ {print $3}')

  2. Extract the list of CredentialsRequest objects from the OKD release image by running the following command:

    $ oc adm release extract \
  --from=$RELEASE_IMAGE \
  --credentials-requests \
  --included \(1)
  --install-config=<path_to_directory_with_installation_configuration>/install-config.yaml \(2)
  --to=<path_to_directory_for_credentials_requests> (3)
    1 The --included parameter includes only the manifests that your specific cluster configuration requires.
    2 Specify the location of the install-config.yaml file.
    3 Specify the path to the directory where you want to store the CredentialsRequest objects. If the specified directory does not exist, this command creates it.

    This command might take a few moments to run.

  3. To enable the ccoctl utility to detect your Azure credentials automatically, log in to the Azure CLI by running the following command:

    $ az login

  4. Use the ccoctl tool to process all CredentialsRequest objects by running the following command:

    $ ccoctl azure create-all \
  --name=<azure_infra_name> \(1)
  --output-dir=<ccoctl_output_dir> \(2)
  --region=<azure_region> \(3)
  --subscription-id=<azure_subscription_id> \(4)
  --credentials-requests-dir=<path_to_credentials_requests_directory> \(5)
  --dnszone-resource-group-name=<azure_dns_zone_resource_group_name> \(6)
  --tenant-id=<azure_tenant_id> (7)
    1 Specify the user-defined name for all created Azure resources used for tracking.
    2 Optional: Specify the directory in which you want the ccoctl utility to create objects. By default, the utility creates objects in the directory in which the commands are run.
    3 Specify the Azure region in which cloud resources will be created.
    4 Specify the Azure subscription ID to use.
    5 Specify the directory containing the files for the component CredentialsRequest objects.
    6 Specify the name of the resource group containing the cluster’s base domain Azure DNS zone.
    7 Specify the Azure tenant ID to use.

    If your cluster uses Technology Preview features that are enabled by the TechPreviewNoUpgrade feature set, you must include the --enable-tech-preview parameter.

    To see additional optional parameters and explanations of how to use them, run the azure create-all --help command.
Verification

  • To verify that the OKD secrets are created, list the files in the <path_to_ccoctl_output_dir>/manifests directory:

    $ ls <path_to_ccoctl_output_dir>/manifests
    Example output
    azure-ad-pod-identity-webhook-config.yaml
cluster-authentication-02-config.yaml
openshift-cloud-controller-manager-azure-cloud-credentials-credentials.yaml
openshift-cloud-network-config-controller-cloud-credentials-credentials.yaml
openshift-cluster-api-capz-manager-bootstrap-credentials-credentials.yaml
openshift-cluster-csi-drivers-azure-disk-credentials-credentials.yaml
openshift-cluster-csi-drivers-azure-file-credentials-credentials.yaml
openshift-image-registry-installer-cloud-credentials-credentials.yaml
openshift-ingress-operator-cloud-credentials-credentials.yaml
openshift-machine-api-azure-cloud-credentials-credentials.yaml

    You can verify that the Microsoft Entra ID service accounts are created by querying Azure. For more information, refer to Azure documentation on listing Entra ID service accounts.

Incorporating the Cloud Credential Operator utility manifests

To implement short-term security credentials managed outside the cluster for individual components, you must move the manifest files that the Cloud Credential Operator utility (ccoctl) created to the correct directories for the installation program.

Prerequisites

  • You have configured an account with the cloud platform that hosts your cluster.

  • You have configured the Cloud Credential Operator utility (ccoctl).

  • You have created the cloud provider resources that are required for your cluster with the ccoctl utility.

Procedure

  1. If you did not set the credentialsMode parameter in the install-config.yaml configuration file to Manual, modify the value as shown:

    Sample configuration file snippet
    apiVersion: v1
baseDomain: example.com
credentialsMode: Manual
# ...

  2. If you used the ccoctl utility to create a new Azure resource group instead of using an existing resource group, modify the resourceGroupName parameter in the install-config.yaml as shown:

    Sample configuration file snippet
    apiVersion: v1
baseDomain: example.com
# ...
platform:
  azure:
    resourceGroupName: <azure_infra_name> (1)
# ...
    1 This value must match the user-defined name for Azure resources that was specified with the --name argument of the ccoctl azure create-all command.

  3. If you have not previously created installation manifest files, do so by running the following command:

    $ openshift-install create manifests --dir <installation_directory>

    where <installation_directory> is the directory in which the installation program creates files.

  4. Copy the manifests that the ccoctl utility generated to the manifests directory that the installation program created by running the following command:

    $ cp /<path_to_ccoctl_output_dir>/manifests/* ./manifests/

  5. Copy the tls directory that contains the private key to the installation directory:

    $ cp -a /<path_to_ccoctl_output_dir>/tls .

Optional: Preparing a private Microsoft Azure cluster for a private image registry

By installing a private image registry on a private Microsoft Azure cluster, you can create private storage endpoints. Private storage endpoints disable public facing endpoints to the registry’s storage account, adding an extra layer of security to your OKD deployment.

Do not install a private image registry on Microsoft Azure Red Hat OpenShift (ARO), because the endpoint can put your Microsoft Azure Red Hat OpenShift cluster in an unrecoverable state.

Use the following guide to prepare your private Microsoft Azure cluster for installation with a private image registry.

Prerequisites

  • You have access to an OKD account with cluster administrator access.

  • You have installed the OpenShift CLI (oc).

  • You have prepared an install-config.yaml that includes the following information:

    • The publish field is set to Internal

  • You have set the permissions for creating a private storage endpoint. For more information, see "Azure permissions for installer-provisioned infrastructure".

Procedure

  1. If you have not previously created installation manifest files, do so by running the following command:

    $ ./openshift-install create manifests --dir <installation_directory>

    This command displays the following messages:

    Example output
    INFO Consuming Install Config from target directory
INFO Manifests created in: <installation_directory>/manifests and <installation_directory>/openshift

  2. Create an image registry configuration object and pass in the networkResourceGroupName, subnetName, and vnetName provided by Microsoft Azure. For example:

    $ touch imageregistry-config.yaml
    apiVersion: imageregistry.operator.openshift.io/v1
kind: Config
metadata:
  name: cluster
spec:
  managementState: "Managed"
  replicas: 2
  rolloutStrategy: RollingUpdate
  storage:
    azure:
      networkAccess:
        internal:
          networkResourceGroupName: <vnet_resource_group> (1)
          subnetName: <subnet_name> (2)
          vnetName: <vnet_name> (3)
        type: Internal
    1 Optional. If you have an existing VNet and subnet setup, replace <vnet_resource_group> with the resource group name that contains the existing virtual network (VNet).
    2 Optional. If you have an existing VNet and subnet setup, replace <subnet_name> with the name of the existing compute subnet within the specified resource group.
    3 Optional. If you have an existing VNet and subnet setup, replace <vnet_name> with the name of the existing virtual network (VNet) in the specified resource group.

    The imageregistry-config.yaml file is consumed during the installation process. If desired, you must back it up before installation.

  3. Move the imageregistry-config.yaml file to the <installation_directory/manifests> folder by running the following command:

    $ mv imageregistry-config.yaml <installation_directory/manifests/>
Next steps

  • After you have moved the imageregistry-config.yaml file to the <installation_directory/manifests> folder and set the required permissions, proceed to "Deploying the cluster".

Additional resources

Deploying the cluster

You can install OKD on a compatible cloud platform.

You can run the create cluster command of the installation program only once, during initial installation.
Prerequisites

  • You have configured an account with the cloud platform that hosts your cluster.

  • You have the OKD installation program and the pull secret for your cluster.

  • You have an Azure subscription ID and tenant ID.

  • If you are installing the cluster using a service principal, you have its application ID and password.

  • If you are installing the cluster using a system-assigned managed identity, you have enabled it on the virtual machine that you will run the installation program from.

  • If you are installing the cluster using a user-assigned managed identity, you have met these prerequisites:

    • You have its client ID.

    • You have assigned it to the virtual machine that you will run the installation program from.

Procedure

  1. Optional: If you have run the installation program on this computer before, and want to use an alternative service principal or managed identity, go to the ~/.azure/ directory and delete the osServicePrincipal.json configuration file.

    Deleting this file prevents the installation program from automatically reusing subscription and authentication values from a previous installation.

  2. Change to the directory that contains the installation program and initialize the cluster deployment:

    $ ./openshift-install create cluster --dir <installation_directory> \ (1)
    --log-level=info (2)
    1 For <installation_directory>, specify the location of your customized ./install-config.yaml file.
    2 To view different installation details, specify warn, debug, or error instead of info.

    If the installation program cannot locate the osServicePrincipal.json configuration file from a previous installation, you are prompted for Azure subscription and authentication values.

  3. Enter the following Azure parameter values for your subscription:

    • azure subscription id: Enter the subscription ID to use for the cluster.

    • azure tenant id: Enter the tenant ID.

  4. Depending on the Azure identity you are using to deploy the cluster, do one of the following when prompted for the azure service principal client id:

    • If you are using a service principal, enter its application ID.

    • If you are using a system-assigned managed identity, leave this value blank.

    • If you are using a user-assigned managed identity, specify its client ID.

  5. Depending on the Azure identity you are using to deploy the cluster, do one of the following when prompted for the azure service principal client secret:

    • If you are using a service principal, enter its password.

    • If you are using a system-assigned managed identity, leave this value blank.

    • If you are using a user-assigned managed identity,leave this value blank.

If previously not detected, the installation program creates an osServicePrincipal.json configuration file and stores this file in the ~/.azure/ directory on your computer. This ensures that the installation program can load the profile when it is creating an OKD cluster on the target platform.

Verification

When the cluster deployment completes successfully:

  • The terminal displays directions for accessing your cluster, including a link to the web console and credentials for the kubeadmin user.

  • Credential information also outputs to <installation_directory>/.openshift_install.log.

Do not delete the installation program or the files that the installation program creates. Both are required to delete the cluster.
Example output
...
INFO Install complete!
INFO To access the cluster as the system:admin user when using 'oc', run 'export KUBECONFIG=/home/myuser/install_dir/auth/kubeconfig'
INFO Access the OpenShift web-console here: https://console-openshift-console.apps.mycluster.example.com
INFO Login to the console with user: "kubeadmin", and password: "password"
INFO Time elapsed: 36m22s

  • The Ignition config files that the installation program generates contain certificates that expire after 24 hours, which are then renewed at that time. If the cluster is shut down before renewing the certificates and the cluster is later restarted after the 24 hours have elapsed, the cluster automatically recovers the expired certificates. The exception is that you must manually approve the pending node-bootstrapper certificate signing requests (CSRs) to recover kubelet certificates. See the documentation for Recovering from expired control plane certificates for more information.

  • It is recommended that you use Ignition config files within 12 hours after they are generated because the 24-hour certificate rotates from 16 to 22 hours after the cluster is installed. By using the Ignition config files within 12 hours, you can avoid installation failure if the certificate update runs during installation.

Logging in to the cluster by using the CLI

You can log in to your cluster as a default system user by exporting the cluster kubeconfig file. The kubeconfig file contains information about the cluster that is used by the CLI to connect a client to the correct cluster and API server. The file is specific to a cluster and is created during OKD installation.

Prerequisites

  • You deployed an OKD cluster.

  • You installed the oc CLI.

Procedure

  1. Export the kubeadmin credentials:

    $ export KUBECONFIG=<installation_directory>/auth/kubeconfig (1)
    1 For <installation_directory>, specify the path to the directory that you stored the installation files in.

  2. Verify you can run oc commands successfully using the exported configuration:

    $ oc whoami
    Example output
    system:admin
Additional resources

Next steps