IBM-Fusion-DF-cli

Installing IBM Fusion Data Foundation using the command line (CLI) and Dynamic storage classes in an Installer Provisioned Infrastructure Openshift cluster.

Overview

Red Hat has this doc which shows how to use the CLI to install and configure Data Foundation using the Local Storage operator and disks from LocalStorage storage classes. This document intends do show how to use dynamic disks ona an IPI cluster, in this case an Openshift cluster installed on VMware.

The pre-requisites

In this example we used Openshift 4.18 and Fusion 2.11. On a fresh installed Openshift cluster on VMware we need to be able to install IBM Fusion Software. In order to proceed we need to get through the pre-requisites listed on Fusion documentation.

To run the steps you need to be logged on the Openshift cluster with the oc command with an administrative user:

oc login -u kubeadmin

And type the password or go to the user name on up right corner of the Openshift web console and got to Copy login command click on Display Token and use:

oc login --token=sha256~<your token> --server=https://api.<your-cluster.your-domain>:6443

The main pre-requisite steps are:

• Obtaining the entitlement key which allows the download of software images from the IBM registry web site. Licensed customers already have access to the entitlement keys, trial version is also available.
• Creating the image pull secret so Openshift can access the IBM Registry.
• Adding IBM Operator Catalog to have all IBM software available on Operators -> Operator Hub

With all the pre-requisites met we can start to configure the Operators

Installing the IBM Fusion Operator

IBM provides documentation on how to install the IBM Fusion Operator using the command line.

The basic steps are to create the deployfusion.yaml file with the content:

apiVersion: v1
kind: Namespace
metadata:
  name: ibm-spectrum-fusion-ns
---
apiVersion: operators.coreos.com/v1
kind: OperatorGroup
metadata:
  name: isf-og
  namespace: ibm-spectrum-fusion-ns
spec:
  targetNamespaces:
  -  ibm-spectrum-fusion-ns
  upgradeStrategy: Default
---
apiVersion: operators.coreos.com/v1alpha1
kind: CatalogSource
metadata:
  name: ibm-operator-catalog
  namespace: openshift-marketplace
spec:
  sourceType: grpc
  image:  icr.io/cpopen/ibm-operator-catalog:latest
  displayName: IBM Operator Catalog
  publisher: IBM
  updateStrategy:
   registryPoll:
    interval: 45m
---
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
  name: isf-operator
  namespace: ibm-spectrum-fusion-ns
spec:
  channel: v2.0
  name: isf-operator
  sourceNamespace: openshift-marketplace
  source: ibm-operator-catalog
  installPlanApproval: Automatic

Apply the file to the cluster:

oc apply -f deployfusion.yaml

Then we need to accept Fusion license, creating the accept_license.yaml file:

apiVersion: prereq.isf.ibm.com/v1
kind: SpectrumFusion
metadata:
  name: spectrumfusion
  namespace: ibm-spectrum-fusion-ns
spec:
  license:
    accept: true

And then apply the file:

oc apply -f accept_license.yaml

Configuring the Data Foundation operator

We need to create the Project/Namespace with namespace.yaml

apiVersion: v1
kind: Namespace
metadata:
 labels:
   openshift.io/cluster-monitoring: "true"
 name: openshift-storage
spec: {}

Then apply namespace.yaml

oc apply -f namespace.yaml

Create the operatorgroup.yaml file

apiVersion: operators.coreos.com/v1
kind: OperatorGroup
metadata:
 name: openshift-storage-operatorgroup
 namespace: openshift-storage
spec:
 targetNamespaces:
 - openshift-storage

Apply operatorgroup.yaml file

oc apply -f operatorgroup.yaml

We need to create the ISF DataFoundation Catalog, creating the isf-datafoundation-catalog.yaml

apiVersion: operators.coreos.com/v1alpha1
kind: CatalogSource
metadata:
 name: isf-data-foundation-catalog
 namespace: openshift-marketplace
spec:
 displayName: Data Foundation Catalog
 image: icr.io/cpopen/isf-data-foundation-catalog:v4.18
 publisher: IBM
 sourceType: grpc
 updateStrategy:
   registryPoll:
     interval: 60m

Apply the isf-datafoundation-catalog.yaml

oc apply -f isf-datafoundation-catalog.yaml

We finally install the Data Foundation operator, with the file subscription.yaml

apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
 name: odf-operator
 namespace: openshift-storage
spec:
 channel: "stable-4.18"
 installPlanApproval: Automatic
 name: odf-operator
 source: isf-data-foundation-catalog
 sourceNamespace: openshift-marketplace

Installing the operator

oc apply -f subscription.yaml

We need to patch the Console operator so the Data Foundation appears on Openshift Web Concole:

oc patch console.operator cluster -n openshift-storage --type json -p '[{"op": "add", "path": "/spec/plugins", "value": ["odf-console"]}]'

Creating the Storage Cluster

With the operators installed we create the ocs-storagecluster.yaml poiting to a Storage Class known by the Openshift, which in this case is VMware.

apiVersion: ocs.openshift.io/v1
kind: StorageCluster
metadata:
 name: ocs-storagecluster
 namespace: openshift-storage
spec:
  arbiter: {}
  encryption:
    kms: {}
  externalStorage: {}
  managedResources:
    cephBlockPools: {}
    cephCluster: {}
    cephConfig: {}
    cephDashboard: {}
    cephFilesystems: {}
    cephNonResilientPools: {}
    cephObjectStoreUsers: {}
    cephObjectStores: {}
    cephToolbox: {}
  mirroring: {}
  monPVCTemplate:
    metadata: {}
    spec:
      accessModes:
      - ReadWriteOnce
      resources:
        requests:
          storage: 50Gi
      storageClassName: thin-csi
      volumeMode: Filesystem
    status: {}
  resources: {}
  storageDeviceSets:
  - config: {}
    count: 1
    dataPVCTemplate:
      metadata: {}
      spec:
        accessModes:
        - ReadWriteOnce
        resources:
          requests:
            storage: 2Ti
        storageClassName: thin-csi
        volumeMode: Block
      status: {}
    name: fusion-storage
    placement: {}
    preparePlacement: {}
    replica: 3
    resources: {}

Then apply the ocs-

oc apply -f ocs-storagecluster.yaml

After all pods are created the Storage Cluster is ready to use.