Prerequisites

If you are running cluster monitoring with an attached PVC for Prometheus, you might experience OOM kills during cluster upgrade. When persistent storage is in use for Prometheus, Prometheus memory usage doubles during cluster upgrade and for several hours after upgrade is complete. To avoid the OOM kill issue, allow worker nodes with double the size of memory that was available prior to the upgrade. For example, if you are running monitoring on the minimum recommended nodes, which is 2 cores with 8 GB of RAM, increase memory to 16 GB. For more information, see BZ#1925061.

About the OpenShift Update Service

The OpenShift Update Service (OSUS) provides over-the-air updates to OpenShift Container Platform, including Red Hat Enterprise Linux CoreOS (RHCOS). It provides a graph, or diagram, that contains the vertices of component Operators and the edges that connect them. The edges in the graph show which versions you can safely update to. The vertices are update payloads that specify the intended state of the managed cluster components.

The Cluster Version Operator (CVO) in your cluster checks with the OpenShift Update Service to see the valid updates and update paths based on current component versions and information in the graph. When you request an update, the CVO uses the release image for that update to upgrade your cluster. The release artifacts are hosted in Quay as container images.

To allow the OpenShift Update Service to provide only compatible updates, a release verification pipeline drives automation. Each release artifact is verified for compatibility with supported cloud platforms and system architectures, as well as other component packages. After the pipeline confirms the suitability of a release, the OpenShift Update Service notifies you that it is available.

The OpenShift Update Service displays all recommended updates for your current cluster. If an upgrade path is not recommended by the OpenShift Update Service, it might be because of a known issue with the update or the target release.

Two controllers run during continuous update mode. The first controller continuously updates the payload manifests, applies the manifests to the cluster, and outputs the controlled rollout status of the Operators to indicate whether they are available, upgrading, or failed. The second controller polls the OpenShift Update Service to determine if updates are available.

Only upgrading to a newer version is supported. Reverting or rolling back your cluster to a previous version is not supported. If your upgrade fails, contact Red Hat support.

During the upgrade process, the Machine Config Operator (MCO) applies the new configuration to your cluster machines. The MCO cordons the number of nodes as specified by the maxUnavailable field on the machine configuration pool and marks them as unavailable. By default, this value is set to 1. The MCO then applies the new configuration and reboots the machine.

If you use Red Hat Enterprise Linux (RHEL) machines as workers, the MCO does not update the kubelet because you must update the OpenShift API on the machines first.

With the specification for the new version applied to the old kubelet, the RHEL machine cannot return to the Ready state. You cannot complete the update until the machines are available. However, the maximum number of unavailable nodes is set to ensure that normal cluster operations can continue with that number of machines out of service.

The OpenShift Update Service is composed of an Operator and one or more application instances.

OpenShift Container Platform upgrade channels and releases

In OpenShift Container Platform 4.1, Red Hat introduced the concept of channels for recommending the appropriate release versions for cluster upgrades. By controlling the pace of upgrades, these upgrade channels allow you to choose an upgrade strategy. Upgrade channels are tied to a minor version of OpenShift Container Platform. For instance, OpenShift Container Platform 4.7 upgrade channels recommend upgrades to 4.7 and upgrades within 4.7. They also recommend upgrades within 4.6 and from 4.6 to 4.7, to allow clusters on 4.6 to eventually upgrade to 4.7. They do not recommend upgrades to 4.8 or later releases. This strategy ensures that administrators explicitly decide to upgrade to the next minor version of OpenShift Container Platform.

Upgrade channels control only release selection and do not impact the version of the cluster that you install; the openshift-install binary file for a specific version of OpenShift Container Platform always installs that version.

OpenShift Container Platform 4.7 offers the following upgrade channels:

  • candidate-4.7

  • fast-4.7

  • stable-4.7

  • eus-4.6 (only available when running 4.6)

candidate-4.7 channel

The candidate-4.7 channel contains candidate builds for a z-stream (4.7.z) and previous minor version releases. Release candidates contain all the features of the product but are not supported. Use release candidate versions to test feature acceptance and assist in qualifying the next version of OpenShift Container Platform. A release candidate is any build that is available in the candidate channel, including ones that do not contain a pre-release version such as -rc in their names. After a version is available in the candidate channel, it goes through more quality checks. If it meets the quality standard, it is promoted to the fast-4.7 or stable-4.7 channels. Because of this strategy, if a specific release is available in both the candidate-4.7 channel and in the fast-4.7 or stable-4.7 channels, it is a Red Hat-supported version. The candidate-4.7 channel can include release versions from which there are no recommended updates in any channel.

You can use the candidate-4.7 channel to upgrade from a previous minor version of OpenShift Container Platform.

Release candidates differ from the nightly builds. Nightly builds are available for early access to features, but updating to or from nightly builds is neither recommended nor supported. Nightly builds are not available in any upgrade channel. You can reference the OpenShift Container Platform release statuses for more build information.

fast-4.7 channel

The fast-4.7 channel is updated with new and previous minor versions of 4.7 as soon as Red Hat declares the given version as a general availability release. As such, these releases are fully supported, are production quality, and have performed well while available as a release candidate in the candidate-4.7 channel from where they were promoted. Some time after a release appears in the fast-4.7 channel, it is added to the stable-4.7 channel. Releases never appear in the stable-4.7 channel before they appear in the fast-4.7 channel.

You can use the fast-4.7 channel to upgrade from a previous minor version of OpenShift Container Platform.

stable-4.7 channel

While the fast-4.7 channel contains releases as soon as their errata are published, releases are added to the stable-4.7 channel after a delay. During this delay, data is collected from Red Hat SRE teams, Red Hat support services, and pre-production and production environments that participate in connected customer program about the stability of the release. You can use the stable-4.7 channel to upgrade from a previous minor version of OpenShift Container Platform.

eus-4.6 channel

In addition to the stable channel, certain minor versions of OpenShift Container Platform offer an Extended Update Support (EUS). These EUS versions extend the maintenance phase for customers with Premium Subscriptions to 14 months. OpenShift Container Platform 4.6 is currently the only minor version with EUS.

Although there is no difference between stable-4.6 and eus-4.6 channels until OpenShift Container Platform 4.6 transitions to the EUS phase, you can switch to the EUS channel as soon as it becomes available. When OpenShift Container Platform 4.6 transitions to the EUS phase of its lifecycle, the stable-4.6 channel will no longer receive subsequent z-stream updates. After you upgrade to a version that is exclusive to the EUS channel, that cluster will no longer be eligible for minor version upgrades until upgrades to the next EUS version become available. The next planned EUS version is to be 4.10 and the upgrade to that version will require a serial set of version upgrades, such as from 4.6 to 4.7 to 4.8 to 4.9 to 4.10.

Additionally, you may only switch to the EUS channel when your cluster is running a supported version of OpenShift Container Platform 4.6.

Finally, if you install a 4.6 version that is exclusive to EUS, you will similarly not be able to upgrade to a later minor version until upgrades are provided to 4.10.

Upgrade version paths

OpenShift Container Platform maintains an upgrade recommendation service that understands the version of OpenShift Container Platform you have installed as well as the path to take within the channel you choose to get you to the next release.

You can imagine seeing the following in the fast-4.7 channel:

  • 4.7.0

  • 4.7.1

  • 4.7.3

  • 4.7.4

The service recommends only upgrades that have been tested and have no serious issues. It will not suggest updating to a version of OpenShift Container Platform that contains known vulnerabilities. For example, if your cluster is on 4.7.1 and OpenShift Container Platform suggests 4.7.4, then it is safe for you to update from 4.7.1 to 4.7.4. Do not rely on consecutive patch numbers. In this example, 4.7.2 is not and never was available in the channel.

Update stability depends on your channel. The presence of an update recommendation in the candidate-4.7 channel does not imply that the update is supported. It means that no serious issues have been found with the update yet, but there might not be significant traffic through the update to suggest stability. The presence of an update recommendation in the fast-4.7 or stable-4.7 channels at any point is a declaration that the update is supported. While releases will never be removed from a channel, update recommendations that exhibit serious issues will be removed from all channels. Updates initiated after the update recommendation has been removed are still supported.

Red Hat will eventually provide supported update paths from any supported release in the fast-4.7 or stable-4.7 channels to the latest release in 4.7.z, although there can be delays while safe paths away from troubled releases are constructed and verified.

Fast and stable channel use and strategies

The fast-4.7 and stable-4.7 channels present a choice between receiving general availability releases as soon as they are available or allowing Red Hat to control the rollout of those updates. If issues are detected during rollout or at a later time, upgrades to that version might be blocked in both the fast-4.7 and stable-4.7 channels, and a new version might be introduced that becomes the new preferred upgrade target.

Customers can improve this process by configuring pre-production systems on the fast-4.7 channel, configuring production systems on the stable-4.7 channel, and participating in the Red Hat connected customer program. Red Hat uses this program to observe the impact of updates on your specific hardware and software configurations. Future releases might improve or alter the pace at which updates move from the fast-4.7 to the stable-4.7 channel.

Restricted network clusters

If you manage the container images for your OpenShift Container Platform clusters yourself, you must consult the Red Hat errata that is associated with product releases and note any comments that impact upgrades. During upgrade, the user interface might warn you about switching between these versions, so you must ensure that you selected an appropriate version before you bypass those warnings.

Switching between channels

A channel can be switched from the web console or through the patch command:

$ oc patch clusterversion version --type json -p '[{"op": "add", "path": "/spec/channel", "value": "<channel>”}]'

The web console will display an alert if you switch to a channel that does not include the current release. The web console does not recommend any updates while on a channel without the current release. You can return to the original channel at any point, however.

Changing your channel might impact the supportability of your cluster. The following conditions might apply:

  • Your cluster is still supported if you change from the stable-4.7 channel to the fast-4.7 channel.

  • You can switch to the candidate-4.7 channel but, some releases for this channel might be unsupported.

  • You can switch from the candidate-4.7 channel to the fast-4.7 channel if your current release is a general availability release.

  • You can always switch from the fast-4.7 channel to the stable-4.7 channel. There is a possible delay of up to a day for the release to be promoted to stable-4.7 if the current release was recently promoted.

Updating a cluster by using the web console

If updates are available, you can update your cluster from the web console.

You can find information about available OpenShift Container Platform advisories and updates in the errata section of the Customer Portal.

Prerequisites
  • Have access to the web console as a user with admin privileges.

Procedure
  1. From the web console, click AdministrationCluster Settings and review the contents of the Details tab.

  2. For production clusters, ensure that the Channel is set to the correct channel for your current minor version, such as stable-4.7.

    For production clusters, you must subscribe to a stable-* or fast-* channel.

    • If the Update status is not Updates available, you cannot upgrade your cluster.

    • Select channel indicates the cluster version that your cluster is running or is updating to.

  3. Select the highest available version and click Save.

    The Input channel Update status changes to Update to <product-version> in progress, and you can review the progress of the cluster update by watching the progress bars for the Operators and nodes.

    If you are upgrading your cluster to the next minor version, like version 4.y to 4.(y+1), it is recommended to confirm your nodes are upgraded before deploying workloads that rely on a new feature. Any pools with worker nodes that are not yet updated are displayed on the Cluster Settings page.

  4. After the update completes and the Cluster Version Operator refreshes the available updates, check if more updates are available in your current channel.

    • If updates are available, continue to perform updates in the current channel until you can no longer update.

    • If no updates are available, change the Channel to the stable-* or fast-* channel for the next minor version, and update to the version that you want in that channel.

    You might need to perform several intermediate updates until you reach the version that you want.

    When you update a cluster that contains Red Hat Enterprise Linux (RHEL) worker machines, those workers temporarily become unavailable during the update process. You must run the upgrade playbook against each RHEL machine as it enters the NotReady state for the cluster to finish updating.

Optional: Adding hooks to perform Ansible tasks on RHEL machines

You can use hooks to run Ansible tasks on the RHEL compute machines during the OpenShift Container Platform update.

About Ansible hooks for upgrades

When you update OpenShift Container Platform, you can run custom tasks on your Red Hat Enterprise Linux (RHEL) nodes during specific operations by using hooks. Hooks allow you to provide files that define tasks to run before or after specific update tasks. You can use hooks to validate or modify custom infrastructure when you update the RHEL compute nodes in you OpenShift Container Platform cluster.

Because when a hook fails, the operation fails, you must design hooks that are idempotent, or can run multiple times and provide the same results.

Hooks have the following important limitations: - Hooks do not have a defined or versioned interface. They can use internal openshift-ansible variables, but it is possible that the variables will be modified or removed in future OpenShift Container Platform releases. - Hooks do not have error handling, so an error in a hook halts the update process. If you get an error, you must address the problem and then start the upgrade again.

Configuring the Ansible inventory file to use hooks

You define the hooks to use when you update the Red Hat Enterprise Linux (RHEL) compute machines, which are also known as worker machines, in the hosts inventory file under the all:vars section.

Prerequisites
  • You have access to the machine that you used to add the RHEL compute machines cluster. You must have access to the hosts Ansible inventory file that defines your RHEL machines.

Procedure
  1. After you design the hook, create a YAML file that defines the Ansible tasks for it. This file must be a set of tasks and cannot be a playbook, as shown in the following example:

    ---
    # Trivial example forcing an operator to acknowledge the start of an upgrade
    # file=/home/user/openshift-ansible/hooks/pre_compute.yml
    
    - name: note the start of a compute machine update
      debug:
          msg: "Compute machine upgrade of {{ inventory_hostname }} is about to start"
    
    - name: require the user agree to start an upgrade
      pause:
          prompt: "Press Enter to start the compute machine update"
  2. Modify the hosts Ansible inventory file to specify the hook files. The hook files are specified as parameter values in the [all:vars] section, as shown:

    Example hook definitions in an inventory file
    [all:vars]
    openshift_node_pre_upgrade_hook=/home/user/openshift-ansible/hooks/pre_node.yml
    openshift_node_post_upgrade_hook=/home/user/openshift-ansible/hooks/post_node.yml

    To avoid ambiguity in the paths to the hook, use absolute paths instead of a relative paths in their definitions.

Available hooks for RHEL compute machines

You can use the following hooks when you update the Red Hat Enterprise Linux (RHEL) compute machines in your OpenShift Container Platform cluster.

Hook name Description

openshift_node_pre_cordon_hook

  • Runs before each node is cordoned.

  • This hook runs against each node in serial.

  • If a task must run against a different host, the task must use delegate_to or local_action.

openshift_node_pre_upgrade_hook

  • Runs after each node is cordoned but before it is updated.

  • This hook runs against each node in serial.

  • If a task must run against a different host, the task must use delegate_to or local_action.

openshift_node_pre_uncordon_hook

  • Runs after each node is updated but before it is uncordoned.

  • This hook runs against each node in serial.

  • If a task must run against a different host, they task must use delegate_to or local_action.

openshift_node_post_upgrade_hook

  • Runs after each node uncordoned. It is the last node update action.

  • This hook runs against each node in serial.

  • If a task must run against a different host, the task must use delegate_to or local_action.

Updating RHEL compute machines in your cluster

After you update your cluster, you must update the Red Hat Enterprise Linux (RHEL) compute machines in your cluster.

Because only Red Hat Enterprise Linux (RHEL) version 7.9 or later is supported for worker (compute) machines, you must not upgrade the RHEL worker machines to version 8.

You can also update your compute machines to another minor version of OpenShift Container Platform if you are using RHEL as the operating system. You do not need to exclude any RPM packages from RHEL when performing a minor version update.

Prerequisites
  • You updated your cluster.

    Because the RHEL machines require assets that are generated by the cluster to complete the update process, you must update the cluster before you update the RHEL worker machines in it.

  • You have access to the local machine that you used to add the RHEL compute machines to your cluster. You must have access to the hosts Ansible inventory file that defines your RHEL machines and the upgrade playbook.

  • For updates to a minor version, the RPM repository is using the same version of OpenShift Container Platform that is running on your cluster.

Procedure
  1. Stop and disable firewalld on the host:

    # systemctl disable --now firewalld.service

    By default, the base OS RHEL with "Minimal" installation option enables firewalld serivce. Having the firewalld service enabled on your host prevents you from accessing OpenShift Container Platform logs on the worker. Do not enable firewalld later if you wish to continue accessing OpenShift Container Platform logs on the worker.

  2. Enable the repositories that are required for OpenShift Container Platform 4.7:

    1. On the machine that you run the Ansible playbooks, update the required repositories:

      # subscription-manager repos --disable=rhel-7-server-ose-4.6-rpms \
                                   --enable=rhel-7-server-ansible-2.9-rpms \
                                   --enable=rhel-7-server-ose-4.7-rpms
    2. On the machine that you run the Ansible playbooks, update the required packages, including openshift-ansible:

      # yum update openshift-ansible openshift-clients
    3. On each RHEL compute node, update the required repositories:

      # subscription-manager repos --disable=rhel-7-server-ose-4.6-rpms \
                                   --enable=rhel-7-server-ose-4.7-rpms  \
                                   --enable=rhel-7-fast-datapath-rpms   \
                                   --enable=rhel-7-server-optional-rpms
  3. Update a RHEL worker machine:

    1. Review the current node status to determine which RHEL worker to update:

      # oc get node
      Example output
      NAME                        STATUS                        ROLES    AGE    VERSION
      mycluster-control-plane-0   Ready                         master   145m   v1.20.0
      mycluster-control-plane-1   Ready                         master   145m   v1.20.0
      mycluster-control-plane-2   Ready                         master   145m   v1.20.0
      mycluster-rhel7-0           NotReady,SchedulingDisabled   worker   98m    v1.14.6+97c81d00e
      mycluster-rhel7-1           Ready                         worker   98m    v1.14.6+97c81d00e
      mycluster-rhel7-2           Ready                         worker   98m    v1.14.6+97c81d00e
      mycluster-rhel7-3           Ready                         worker   98m    v1.14.6+97c81d00e

      Note which machine has the NotReady,SchedulingDisabled status.

    2. Review your Ansible inventory file at /<path>/inventory/hosts and update its contents so that only the machine with the NotReady,SchedulingDisabled status is listed in the [workers] section, as shown in the following example:

      [all:vars]
      ansible_user=root
      #ansible_become=True
      
      openshift_kubeconfig_path="~/.kube/config"
      
      [workers]
      mycluster-rhel7-0.example.com
    3. Change to the openshift-ansible directory:

      $ cd /usr/share/ansible/openshift-ansible
    4. Run the upgrade playbook:

      $ ansible-playbook -i /<path>/inventory/hosts playbooks/upgrade.yml (1)
      1 For <path>, specify the path to the Ansible inventory file that you created.

      The upgrade playbook only upgrades the OpenShift Container Platform packages. It does not update the operating system packages.

  4. Follow the process in the previous step to update each RHEL worker machine in your cluster.

  5. After you update all of the workers, confirm that all of your cluster nodes have updated to the new version:

    # oc get node
    Example output
    NAME                        STATUS                        ROLES    AGE    VERSION
    mycluster-control-plane-0   Ready                         master   145m   v1.20.0
    mycluster-control-plane-1   Ready                         master   145m   v1.20.0
    mycluster-control-plane-2   Ready                         master   145m   v1.20.0
    mycluster-rhel7-0           NotReady,SchedulingDisabled   worker   98m    v1.20.0
    mycluster-rhel7-1           Ready                         worker   98m    v1.20.0
    mycluster-rhel7-2           Ready                         worker   98m    v1.20.0
    mycluster-rhel7-3           Ready                         worker   98m    v1.20.0
  6. Optional: Update the operating system packages that were not updated by the upgrade playbook. To update packages that are not on 4.7, use the following command:

    # yum update

    You do not need to exclude RPM packages if you are using the same RPM repository that you used when you installed 4.7.