Installing and setting up Multicloud Object Gateway for IBM Cloud Pak for Data

To install and configure Multicloud Object Gateway:

1. Follow the appropriate documentation to install Multicloud Object Gateway:
   Remember: Skip this step if you use Red Hat OpenShift Data Foundation or IBM Storage Fusion Data Foundation for persistent storage.

   Storage provider	Documentation links
   Red Hat OpenShift Data Foundation	Refer to the Product Documentation for Red Hat OpenShift Data Foundation: Version 4.12 Version 4.14 Version 4.15 See Deploy standalone Multicloud Object Gateway in the Deploying section for your cloud environment. Important: When you install Multicloud Object Gateway, use an RWO storage class or point to external, enterprise-grade S3 compatible storage.
   IBM Storage Fusion Data Foundation	Complete the appropriate tasks for your environment: IBM Storage Fusion Complete Deploying IBM Storage Fusion in the IBM Storage Fusion documentation: Version 2.6 Version 2.7 Version 2.8 Complete Storage cluster creation in the IBM Storage Fusion documentation: Version 2.6 Version 2.7 Version 2.8 Important: When you install Multicloud Object Gateway, use an RWO storage class or point to external, enterprise-grade S3 compatible storage. IBM Storage Fusion HCI System Complete Deploying stand-alone Multicloud Object Gateway on IBM Storage Fusion HCI System for hosting S3 object in the IBM Storage Fusion HCI System documentation: Version 2.6 Version 2.7 Version 2.8

2. Best practice If the backing store that was created during the Multicloud Object Gateway installation uses pv-pool, create a backing store on enterprise-grade S3 compatible storage or create a backing store for each service that you plan to install.

   If your backing store points to enterprise-grade S3-compatible storage, skip this step.

   Storage provider	Documentation links
   Red Hat OpenShift Data Foundation	Refer to the Product Documentation for Red Hat OpenShift Data Foundation: Version 4.12 Version 4.14 Version 4.15 See Multicloud Object Gateway > Creating a new backing store in the Deploying section for your cloud environment.
   IBM Storage Fusion Data Foundation	See Creating a new backing store in the IBM Storage Fusion documentation: Version 2.6 Version 2.7 Version 2.8

3. Optional: If you want to create different IDs for each service that will access Multicloud Object Gateway, create a NooBaa account for each service.
   Repeat the following steps for each account that you want to create:
   1. Set the ACCOUNT_NAME environment variable to the name that you want to use for the service account.

      For example, if you are creating an account for watsonx Assistant, you might use watson-assistant-noobaa-account.

      export ACCOUNT_NAME=<service-account-name>

   2. Set the BACKING_STORE environment variable to the name of the backing store that you will use for the service.

      If you created multiple backing stores, ensure that you specify the correct backing store.

      export BACKING_STORE=<backing-store-name>

   3. Create the NooBaa account:

      cat <<EOF | oc apply -f -
      apiVersion: noobaa.io/v1alpha1
      kind: NooBaaAccount
      metadata:
        labels:
          app: noobaa
        name: ${ACCOUNT_NAME}
        namespace: openshift-storage
      spec:
        default_resource: ${BACKING_STORE}
        allow_bucket_creation: true
        allowed_buckets:
          full_permission: true
          permission_list: []
      EOF