5. Configure the Object Storage Plugin¶

You are going to install the IBM Cloud Object Storage Plugin in your cluster, using the Helm CLI tool in this section.

Add a Helm repository where IBM Cloud Object Storage Plugin chart resides.

helm repo add ibm-charts https://icr.io/helm/ibm-charts

outputs,

$ helm repo add ibm-charts https://icr.io/helm/ibm-charts

`ibm-charts` has been added to your repositories

Refresh your local Helm repository.

helm repo update

outputs,

$ helm repo update

Hang tight while we grab the latest from your chart repositories...
...Successfully got an update from the "ibm-charts" chart repository
Update Complete. ⎈ Happy Helming!⎈

Download and unzip the IBM Cloud Object Storage plugin to your client, then install the plugin to your cluster from local client.

helm pull --untar ibm-charts/ibm-object-storage-plugin
ls -al
helm plugin install ./ibm-object-storage-plugin/helm-ibmc

should result in,

$ helm plugin install ./ibm-object-storage-plugin/helm-ibmc

Installed plugin: ibmc

Housekeeping to allow execution of the ibmc.sh script by making the file executable.
```
chmod 755 $HOME/.local/share/helm/plugins/helm-ibmc/ibmc.sh
```
Verify the IBM Cloud Object Storage plugin installation. The plugin usage information should be displayed when running the command below.
```
helm ibmc --help
```
Before using the IBM Cloud Object Storage Plugin, configuration changes are required.
In the Cloud Shell where you downloaded the IBM Cloud Object Storage plugin, navigate to the /project/cos-with-s3fs/ibm-object-storage-plugin/templates folder of the IBM Cloud Object Storage Plugin installation.
```
ls -al ibm-object-storage-plugin/templates
```
Make sure the provisioner-sa.yaml file is present and configure it to access the COS service using the COS service credentials secret cos-write-access that you created in the previous section.
In the Theia editor, click File > Open, and browse to the /project/cos-with-s3fs/ibm-object-storage-plugin/templates directory and open the file provisioner-sa.yaml.
Search for content ibmcloud-object-storage-secret-reader in the file around line 62.

in a vi editor, - Type colon : - Type /ibmcloud-object-storage-secret-reader - Press <ENTER> key
Scroll a few lines down to line #72 and find the section that is commented out #resourceNames: [""].
```
rules:
- apiGroups: [""]
    resources: ["secrets"]
    #resourceNames: [""]
```
Uncomment the line and change the section to set the secret to cos-write-access and allow access to the COS instance,
```
rules:
- apiGroups: [""]
    resources: ["secrets"]
    resourceNames: ["cos-write-access"]
```
Save the change and close the file.

Install the configured storage classes for IBM Cloud Object Storage, which will use the edited template file.

helm ibmc install ibm-object-storage-plugin ./ibm-object-storage-plugin

outputs,

$ helm ibmc install ibm-object-storage-plugin ./ibm-object-storage-plugin

Helm version: v3.2.0+ge11b7ce
Installing the Helm chart...
PROVIDER: CLASSIC
DC: hou02
Chart: ./ibm-object-storage-plugin
NAME: ibm-object-storage-plugin
LAST DEPLOYED: Sat May 23 17:45:25 2020
NAMESPACE: default
STATUS: deployed
REVISION: 1
NOTES:
Thank you for installing: ibm-object-storage-plugin.   Your release is named: ibm-object-storage-plugin
....
<and a whole lot more instructions>

Verify that the storage classes are created successfully.

oc get storageclass | grep 'ibmc-s3fs'

outputs,

$ oc get storageclass | grep 'ibmc-s3fs'

ibmc-s3fs-cold-cross-region    ibm.io/ibmc-s3fs    19s
ibmc-s3fs-cold-regional    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-flex-cross-region    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-flex-perf-cross-region    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-flex-perf-regional    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-flex-regional    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-standard-cross-region    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-standard-perf-cross-region    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-standard-perf-regional    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-standard-regional    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-vault-cross-region    ibm.io/ibmc-s3fs   19s
ibmc-s3fs-vault-regional    ibm.io/ibmc-s3fs   19s

Review the storage class ibmc-s3fs-standard-regional.

oc describe storageclass ibmc-s3fs-standard-regional

outputs,

$ oc describe storageclass ibmc-s3fs-standard-regional

Name:                  ibmc-s3fs-standard-regional
IsDefaultClass:        No
Annotations:           meta.helm.sh/release-name=ibm-object-storage-plugin,meta.helm.sh/release-namespace=default
Provisioner:           ibm.io/ibmc-s3fs
Parameters:            ibm.io/chunk-size-mb=16,ibm.io/curl-debug=false,ibm.io/debug-level=warn,ibm.io/iam-endpoint=https://iam.bluemix.net,ibm.io/kernel-cache=true,ibm.io/multireq-max=20,ibm.io/object-store-endpoint=NA,ibm.io/object-store-storage-class=NA,ibm.io/parallel-count=2,ibm.io/s3fs-fuse-retry-count=5,ibm.io/stat-cache-size=100000,ibm.io/tls-cipher-suite=AESGCM
AllowVolumeExpansion:  <unset>
MountOptions:          <none>
ReclaimPolicy:         Delete
VolumeBindingMode:     Immediate
Events:                <none>

Additional information is available at https://cloud.ibm.com/docs/containers?topic=containers-object_storage#configure_cos.

Verify that plugin pods are in "Running" state and indicate READY state of 1/1:
```
oc get pods -n kube-system -o wide | grep object
```
outputs,
```
$ oc get pods -n kube-system -o wide | grep object

ibmcloud-object-storage-driver-p4ljp             0/1     Running   0          32s     10.169.231.148   10.169.231.148   <none>           <none>
ibmcloud-object-storage-driver-zqb4h             0/1     Running   0          32s     10.169.231.153   10.169.231.153   <none>           <none>
ibmcloud-object-storage-plugin-fbb867887-msqcg   0/1     Running   0          32s     172.30.136.24    10.169.231.153   <none>           <none>
```
If the pods are not READY and indicate 0/1 then wait and re-run the command until the READY state says 1/1.

The installation is successful when one ibmcloud-object-storage-plugin pod and one or more ibmcloud-object-storage-driver pods are in running state.

The number of ibmcloud-object-storage-driver pods equals the number of worker nodes in your cluster. All pods must be in a Running state for the plug-in to function properly. If the pods fail, run kubectl describe pod -n kube-system <pod_name> to find the root cause for the failure.

Next¶

6. Create the PersistentVolumeClaim