Applying fix packs between major releases

Mirroring fix pack images (air-gapped installations only)

In the applicable topic (Mirroring images with a bastion host or Mirroring images with a portable compute or storage device (file system)):

1. Follow the steps in the section, "2. Set environment variables and download CASE files".

2. Continue with "3. Mirror the images". Complete all steps from 3.1 to 3.4. (Don't complete "3.5. Configure the OpenShift cluster".)

Applying fix packs on OpenShift installations

Applying the latest fix pack to catalog sources

For OpenShift clusters that are online, update the catalog sources and images by applying the latest 16.1.3 fix pack.

• If you are using specific catalog sources for each operator (not the IBM Operator Catalog), follow the procedure in Adding catalog sources to an OpenShift cluster. The commands on this page are updated every time a new fix pack is released, so rerunning the oc apply command applies the latest fix packs.

• If you are using the IBM Operator Catalog, no action is needed, because the IBM Operator Catalog polls every 45 minutes.

  Tip: Applying individual catalog sources (using specific catalog sources for each operator, rather than the IBM Operator Catalog) is the most effective way to fully control software versioning on an OpenShift cluster.

For OpenShift clusters that are disconnected or air-gapped, go to the section, "Adding specific catalog sources for each operator" in Adding catalog sources to an OpenShift cluster. Follow the steps to log in and apply the catalog sources.

Applying fix packs to operators

Applying the fix pack to the operator for the current release involves updating to the latest operator channel.

1. For each operator, find the corresponding operator channel that is listed in Operator and instance versions for this release.

2. Update your operators to the latest 16.1.3 fix pack release. Begin by confirming the channel to which each operator is subscribed. Follow steps 1-5 in Upgrading operators by using the OpenShift web console or steps 1-4 in Upgrading operators using the CLI.

   • If the fix pack is on the latest (correct) channel, do one of the following:

     • Automatic approval is configured - No need to do anything; the fix is applied automatically when the catalog source is updated.

     • Manual approval is configured (not recommended) - Accept an install plan.

   • If the fix pack is on a different channel, change the subscription to target the latest channel that is part of this release. Follow steps 5-7 in Upgrading operators by using the OpenShift web console or steps 5-6 in Upgrading operators using the CLI.

Applying fix packs to operators (AKS installations only)

1. Use the IBM Helm chart repository

   • If you have already added the IBM Helm chart to your local repository, run the following command to update it:

     helm repo update ibm-helm

   • If you have not already added the IBM Helm chart to your local repository, run the following command:

     helm repo add ibm-helm https://raw.githubusercontent.com/IBM/charts/master/repo/ibm-helm

2. Download the Helm chart archive for the latest fix pack and extract the contents to a temporary location (for example, /tmp).

   export TEMP_LOCATION=/tmp
   helm pull ibm-helm/ibm-integration --version 8.1.x --untar --untardir ${TEMP_LOCATION}

   Your chart artifacts should be extracted to a subdirectory with the name /tmp/ibm-integration.

3. Update the custom resource definitions (CRDs). Before updating, validate that a newer version of the operator was not previously installed on the cluster. If the CRDs on the cluster are newer than the ones for this fix pack, the following command will downgrade them.

   kubectl replace -f ${TEMP_LOCATION}/ibm-integration/crds/*

4. Run the following command to specify the namespace that was used for the original installation. In the following example, integration is the namespace name.

   kubectl config set-context --current --namespace=integration

5. The upgrade must use the same installation mode that was used for the original installation. Confirm the installation mode by running this command:

   helm get values ibm-integration

   In the response, review the value for installMode (this value defaults to OwnNamespace if not set). For example:

   USER-SUPPLIED VALUES:
   operator:
     installMode: AllNamespaces

6. Upgrade the operator by running one of the following commands, depending on which installation mode was used for the original installation.

   • All namespaces on the cluster mode

     helm upgrade ibm-integration --set operator.installMode="AllNamespaces" ${TEMP_LOCATION}/ibm-integration

   • A single namespace on the cluster mode:

     helm upgrade ibm-integration ${TEMP_LOCATION}/ibm-integration

7. Apply fix packs to the operators of instances in Cloud Pak for Integration. For instructions on applying fix packs to operators that are installed on AKS, see the applicable product documentation:

   • IBM App Connect operator

   • IBM API Connect operator

   • IBM DataPower Gateway operator

   • IBM Event Streams operator

   • IBM Event Endpoint Management operator

   • IBM Event Processing operator

Applying fix packs to instances

If the spec.version of your Kubernetes resource is set to a channel, the instance is automatically updated to the latest fix pack when the operator is updated (as described in the previous section).

• By using the Platform UI: After upgrading the operator, you get an "Upgrade available" link. Upgrade the instance by using the Upgrade panel, following the procedure for "Upgrading instances on the Integration versions and upgrades page" in Managing versions and upgrades by using the Platform UI.

• By using the CLI: Edit the Kubernetes resources by updating the spec.version to the latest version. You can find the latest version in the CR (the latest value given for status.versions.availableVersions), or by going to Operator and instance versions for this release and finding the version in the "Latest instance" column.