Installing the IBM Cloud Object Storage plug-in

Virtual Private Cloud Classic infrastructure

Install the IBM Cloud Object Storage plug-in with a Helm chart to set up pre-defined storage classes for IBM Cloud Object Storage. You can use these storage classes to create a PVC to provision IBM Cloud Object Storage for your apps.

If you are migrating from RHEL 7 to RHEL 8, you must uninstall and then reinstall the plug-in version 2.2.6 or later. If you are upgrading from a chart version before 2.2.5, you must uninstall and reinstall the plug-in and then re-create your PVCs and pods or the upgrade will fail.

Prerequisites: The IBM Cloud Object Storage plug-in requires at least 0.2 vCPU and 128 MB of memory.

Before you begin: Log in to your account. If applicable, target the appropriate resource group. Set the context for your cluster.

To install the ibmc Helm plug-in and the ibm-object-storage-plugin:

Make sure that your worker node applies the latest patch for your minor version to run your worker node with the latest security settings. The patch version also ensures that the root password on the worker node is renewed.

If you did not apply updates or reload your worker node within the last 90 days, your root password on the worker node expires and the installation of the storage plug-in might fail.
1. List the current patch version of your worker nodes.
```
ibmcloud ks worker ls --cluster <cluster_name_or_ID>
```
  Example output
```
OK
ID                                                  Public IP        Private IP     Machine Type           State    Status   Zone    Version
kube-dal10-crb1a23b456789ac1b20b2nc1e12b345ab-w26   169.xx.xxx.xxx    10.xxx.xx.xxx   b3c.4x16.encrypted     normal   Ready    dal10   1.29_1523*
```
  If your worker node does not apply the latest patch version, you see an asterisk (*) in the Version column of your CLI output.
2. Review the Kubernetes version information to find the latest changes.
3. Apply the latest patch version by reloading your worker node. Follow the instructions in the ibmcloud ks worker reload command to safely reschedule any running pods on your worker node before you reload your worker node. Note that during the reload, your worker node machine is updated with the latest image and data is deleted if not stored outside the worker node.
Review the change log and verify support for your cluster version and architecture.
Follow the instructions to install the version 3 Helm client on your local machine.

If you enabled VRF and service endpoints in your IBM Cloud account, you can use the private IBM Cloud Helm repository to keep your image pull traffic on the private network. If you can't enable VRF or service endpoints in your account, use the public Helm repository.

Add the IBM Cloud Helm repo to your cluster.

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

Update the Helm repos to retrieve the most recent version of all Helm charts in this repo.
```
helm repo update
```
If you installed the IBM Cloud Object Storage Helm plug-in earlier, remove the ibmc plug-in.
```
helm plugin uninstall ibmc
```
Download the Helm charts and unpack the charts in your current directory.
```
helm fetch --untar ibm-helm/ibm-object-storage-plugin
```
If the output shows Error: failed to untar: a file or directory with the name ibm-object-storage-plugin already exists, delete your ibm-object-storage-plugin directory and rerun the helm fetch command.

If you use OS X or a Linux distribution, install the IBM Cloud Object Storage Helm plug-in ibmc. The plug-in automatically retrieves your cluster location and to set the API endpoint for your IBM Cloud Object Storage buckets in your storage classes. If you use Windows as your operating system, continue with the next step.

Install the Helm plug-in.

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

Verify that the ibmc plug-in is installed successfully.

helm ibmc --help

Example output

Helm version: v3.13.1+g3547a4b
Install or upgrade Helm charts in IBM K8S Service(IKS)
Usage:
helm ibmc [command]
Available Commands:
install           Install a Helm chart
upgrade           Upgrade the release to a new version of the Helm chart
Available Flags:
-h, --help        (Optional) This text.
-u, --update      (Optional) Update this plugin to the latest version
Example Usage:
    Install: helm ibmc install ibm-object-storage-plugin ibm-helm/ibm-object-storage-plugin
    Upgrade: helm ibmc upgrade [RELEASE] ibm-helm/ibm-object-storage-plugin
Note:
    1. It is always recommended to install latest version of ibm-object-storage-plugin chart.
    2. It is always recommended to have 'kubectl' client up-to-date.

Optional: If the output shows the error Error: fork/exec /home/iksadmin/.helm/plugins/helm-ibmc/ibmc.sh: permission denied, run the following command.
```
chmod 755 /Users/<user_name>/Library/helm/plugins/helm-ibmc/ibmc.sh
```
Then retry the ibmc --help command.
```
helm ibmc --help
```

Optional: Limit the IBM Cloud Object Storage plug-in to access only the Kubernetes secrets that hold your IBM Cloud Object Storage service credentials. By default, the plug-in can access all Kubernetes secrets in your cluster.
1. Create your IBM Cloud Object Storage service instance.
2. Store your IBM Cloud Object Storage service credentials in a Kubernetes secret).
3. Change directories to the ibm-object-storage-plugin directory.
```
cd ibm-object-storage-plugin
```
4. From the ibm-object-storage-plugin, navigate to the templates directory and list available files. OS X and Linux
```
cd templates && ls
```
  Windows
```
chdir templates && dir
```
5. Open the provisioner-sa.yaml file and look for the ibmcloud-object-storage-secret-reader ClusterRole definition.
6. Add the name of the secret that you created earlier to the list of secrets that the plug-in is authorized to access in the resourceNames section.
```
kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1beta1
metadata:
 name: ibmcloud-object-storage-secret-reader
rules:
- apiGroups: [""]
  resources: ["secrets"]
  resourceNames: ["<secret_name1>","<secret_name2>"]
  verbs: ["get"]
```
7. Save your changes and navigate to your working directory.
Install the ibm-object-storage-plugin in your cluster. When you install the plug-in, pre-defined storage classes are added to your cluster. If you completed the previous step for limiting the IBM Cloud Object Storage plug-in to access only the Kubernetes secrets that hold your IBM Cloud Object Storage service credentials and you are still targeting the templates directory, change directories to your working directory. To set a limit on how much storage is available for the bucket, set the --set quotaLimit=true VPC clusters only: To enable authorized IPs on VPC, set the --set bucketAccessPolicy=true option.

If you don't set the --set quotaLimit=true option during installation, you can't set quotas for your PVCs.

Example helm ibmc install commands for OS X and Linux

helm ibmc install ibm-object-storage-plugin ibm-helm/ibm-object-storage-plugin --set license=true [--set quotaLimit=true/false] [--set bucketAccessPolicy=false] [--set allowCrossNsSecret=true/false]

quotaLimit: A quota limit sets the maximum amount of storage (in bytes) available for a bucket. If you set this option to true, then when you create PVCs, the quota on buckets created by those PVCs is equal to the PVC size. The default value is true.
allowCrossNsSecret: By default, the plug-in searches for the Kubernetes secret in namespaces other than the PVC namespace. If you set this option to false the plug-in searches for the Kubernetes secret in only the PVC namespace. The default value is true.

Example helm install command for Windows.

helm install ibm-object-storage-plugin ./ibm-object-storage-plugin --set dcname="${DC_NAME}" --set provider="${CLUSTER_PROVIDER}" --set workerOS="${WORKER_OS}" --region="${REGION} --set platform="${PLATFORM}" --set license=true [--set bucketAccessPolicy=false]

DC_NAME: The cluster data center. To retrieve the data center, run kubectl get cm cluster-info -n kube-system -o jsonpath="{.data.cluster-config\.json}{'\n'}". Store the data center value in an environment variable by running SET DC_NAME=<datacenter>. Optional: Set the environment variable in Windows PowerShell by running $env:DC_NAME="<datacenter>".
CLUSTER_PROVIDER: The infrastructure provider. To retrieve this value, run kubectl get nodes -o jsonpath="{.items[*].metadata.labels.ibm-cloud\.kubernetes\.io\/iaas-provider}{'\n'}". If the output from the previous step contains softlayer, then set the CLUSTER_PROVIDER to "IBMC". If the output contains gc, ng, or g2, then set the CLUSTER_PROVIDER to "IBMC-VPC". Store the infrastructure provider in an environment variable. For example: SET CLUSTER_PROVIDER="IBMC-VPC".
WORKER_OS and PLATFORM: The operating system of the worker nodes. To retrieve these values, run kubectl get nodes -o jsonpath="{.items[*].metadata.labels.ibm-cloud\.kubernetes\.io\/os}{'\n'}". Store the operating system of the worker nodes in an environment variable. For IBM Cloud Kubernetes Service clusters, run SET WORKER_OS="debian" and SET PLATFORM="k8s".
REGION: The region of the worker nodes. To retrieve this value, run kubectl get nodes -o yaml | grep 'ibm-cloud\.kubernetes\.io/region'. Store the region of the worker nodes in an environment variable by running SET REGION="< region>".

Updating the IBM Cloud Object Storage plug-in

You can upgrade the existing IBM Cloud Object Storage plug-in to the most recent version.

Get the name of your IBM Cloud Object Storage plug-in Helm release and the version of the plug-in in your cluster.

helm ls -A | grep object

Example output

NAME              NAMESPACE      REVISION    UPDATED                                   STATUS      CHART                                  APP VERSION               
<release_name>  <namespace>     1           2020-02-13 16:05:58.599679 -0500 EST    deployed    ibm-object-storage-plugin-1.1.2        1.1.2

Update the IBM Cloud Helm repo to retrieve the most recent version of all Helm charts in this repo.
```
helm repo update
```
Update the IBM Cloud Object Storage ibmc Helm plug-in to the most recent version.
```
helm ibmc --update
```
Install the most recent version of the ibm-object-storage-plugin for your operating system.

Example helm ibmc upgrade command for OS X and Linux.
```
helm ibmc upgrade ibm-object-storage-plugin ibm-helm/ibm-object-storage-plugin --set license=true [--set bucketAccessPolicy=false]
```
Example helm upgrade command for Windows.
```
helm upgrade ibm-object-storage-plugin ./ibm-object-storage-plugin --set dcname="${DC_NAME}" --set provider="${CLUSTER_PROVIDER}" --set workerOS="${WORKER_OS}" --region="${REGION} --set platform="${PLATFORM}" --set license=true [--set bucketAccessPolicy=false]
```
DC_NAME

The cluster data center. To retrieve the data center, run kubectl get cm cluster-info -n kube-system -o jsonpath="{.data.cluster-config\.json}{'\n'}". Store the data center value in an environment variable by running SET DC_NAME=<datacenter>. Optional: Set the environment variable in Windows PowerShell by running $env:DC_NAME="<datacenter>".

CLUSTER_PROVIDER

The infrastructure provider. To retrieve this value, run kubectl get nodes -o jsonpath="{.items[*].metadata.labels.ibm-cloud\.kubernetes\.io\/iaas-provider}{'\n'}". If the output from the previous step contains softlayer, then set the CLUSTER_PROVIDER to "IBMC". If the output contains gc, ng, or g2, then set the CLUSTER_PROVIDER to "IBMC-VPC". Store the infrastructure provider in an environment variable. For example: SET CLUSTER_PROVIDER="IBMC-VPC".

WORKER_OS and PLATFORM

The operating system of the worker nodes. To retrieve these values, run kubectl get nodes -o jsonpath="{.items[*].metadata.labels.ibm-cloud\.kubernetes\.io\/os}{'\n'}". Store the operating system of the worker nodes in an environment variable. For IBM Cloud Kubernetes Service clusters, run SET WORKER_OS="debian" and SET PLATFORM="k8s".

REGION

The region of the worker nodes. To retrieve this value, run kubectl get nodes -o yaml | grep 'ibm-cloud\.kubernetes\.io/region'. Store the region of the worker nodes in an environment variable by running SET REGION="< region>". |
Verify that the ibmcloud-object-storage-plugin is successfully upgraded. The upgrade of the plug-in is successful when you see deployment "ibmcloud-object-storage-plugin" successfully rolled out in your CLI output.
```
kubectl rollout status deployment/ibmcloud-object-storage-plugin -n ibm-object-s3fs
```
Verify that the ibmcloud-object-storage-driver is successfully upgraded. The upgrade is successful when you see daemon set "ibmcloud-object-storage-driver" successfully rolled out in your CLI output.
```
kubectl rollout status ds/ibmcloud-object-storage-driver -n ibm-object-s3fs
```
Verify that the IBM Cloud Object Storage pods are in a Running state.
```
kubectl get pods -n <namespace> -o wide | grep object-storage
```

If you're having trouble updating the IBM Cloud Object Storage plug-in, see Object storage: Installing the Object storage ibmc Helm plug-in fails and Object storage: Installing the IBM Cloud Object Storage plug-in fails.

Removing the IBM Cloud Object Storage plug-in

If you don't want to provision and use IBM Cloud Object Storage in your cluster, you can uninstall the ibm-object-storage-plugin and the ibmc Helm plug-in.

Removing the ibmc Helm plug-in or the ibm-object-storage-plugin doesn't remove existing PVCs, PVs, or data. When you remove the ibm-object-storage-plugin, all the related driver pods and daemon sets are removed from your cluster, which means you can't provision new IBM Cloud Object Storage for your cluster unless you configure your app to use the IBM Cloud Object Storage API directly. There is no impact on existing PVCs and PVs.

Before you begin:

Log in to your account. If applicable, target the appropriate resource group. Set the context for your cluster.
Make sure that you don't have any PVCs or PVs in your cluster that use IBM Cloud Object Storage. To list all pods that mount a specific PVC, run kubectl get pods --all-namespaces -o=jsonpath='{range .items[*]}{"\n"}{.metadata.name}{":\t"}{range .spec.volumes[*]}{.persistentVolumeClaim.claimName}{" "}{end}{end}' | grep "<pvc_name>".

To remove the ibmc Helm plug-in and the ibm-object-storage-plugin:

Get the name of your ibm-object-storage-plugin Helm installation.

helm ls -A | grep ibm-object-storage-plugin

Example output

NAME                         NAMESPACE    REVISION    UPDATED                                 STATUS      CHART                                  APP VERSION               
ibm-object-storage-plugin    default      2           2020-04-01 08:46:01.403477 -0400 EDT    deployed    ibm-object-storage-plugin-1.1.4        1.1.4

Uninstall the ibm-object-storage-plugin.
```
helm uninstall <release_name>
```
Example command for a release named ibm-object-storage-plugin.
```
helm uninstall ibm-object-storage-plugin
```
Verify that the ibm-object-storage-plugin pods are removed.
```
kubectl get pod -n <namespace> | grep object-storage
```
The removal of the pods is successful if no pods are displayed in your CLI output. The removal of the storage classes is successful if no storage classes are displayed in your CLI output.
Verify that the storage classes are removed.
```
kubectl get storageclasses | grep s3
```
If you use OS X or a Linux distribution, remove the ibmc Helm plug-in. If you use Windows, this step is not required.
1. Remove the ibmc Helm plug-in.
```
helm plugin uninstall ibmc
```
2. Verify that the ibmc plug-in is removed. The ibmc plug-in is removed successfully if the ibmc plug-in is not listed in your CLI output.
```
helm plugin list
```
  Example output
```
NAME    VERSION    DESCRIPTION
```

Deciding on the object storage configuration

IBM Cloud Kubernetes Service provides pre-defined storage classes that you can use to create buckets with a specific configuration.

List available storage classes in IBM Cloud Kubernetes Service.

kubectl get storageclasses | grep s3

Example output

ibmc-s3fs-cold-cross-region            ibm.io/ibmc-s3fs   8m
ibmc-s3fs-cold-regional                ibm.io/ibmc-s3fs   8m
ibmc-s3fs-smart-cross-region            ibm.io/ibmc-s3fs   8m
ibmc-s3fs-smart-perf-cross-region       ibm.io/ibmc-s3fs   8m
ibmc-s3fs-smart-perf-regional           ibm.io/ibmc-s3fs   8m
ibmc-s3fs-smart-regional                ibm.io/ibmc-s3fs   8m
ibmc-s3fs-standard-cross-region        ibm.io/ibmc-s3fs   8m
ibmc-s3fs-standard-perf-cross-region   ibm.io/ibmc-s3fs   8m
ibmc-s3fs-standard-perf-regional       ibm.io/ibmc-s3fs   8m
ibmc-s3fs-standard-regional            ibm.io/ibmc-s3fs   8m
ibmc-s3fs-vault-cross-region           ibm.io/ibmc-s3fs   8m
ibmc-s3fs-vault-regional               ibm.io/ibmc-s3fs   8m

Choose a storage class that fits your data access requirements. The storage class determines the pricing for storage capacity, read and write operations, and outbound bandwidth for a bucket. The option that is correct for you is based on how frequently data is read and written to your service instance.
- Standard: This option is used for hot data that is accessed frequently. Common use cases are web or mobile apps.
- Vault: This option is used for workloads or cool data that are accessed infrequently, such as once a month or less. Common use cases are archives, short-term data retention, digital asset preservation, tape replacement, and disaster recovery.
- Cold: This option is used for cold data that is rarely accessed (every 90 days or less), or inactive data. Common use cases are archives, long-term backups, historical data that you keep for compliance, or workloads and apps that are rarely accessed.
- Smart: This option is used for workloads and data that don't follow a specific usage pattern, or that are too huge to determine or predict a usage pattern.
Decide on the level of resiliency for the data that is stored in your bucket.
- Cross-region: With this option, your data is stored across three regions within a geolocation for highest availability. If you have workloads that are distributed across regions, requests are routed to the nearest regional endpoint. The API endpoint for the geolocation is automatically set by the ibmc Helm plug-in that you installed earlier based on the location that your cluster is in. For example, if your cluster is in US South, then your storage classes are configured to use the US GEO API endpoint for your buckets. For more information, see Regions and endpoints.
- Regional: With this option, your data is replicated across multiple zones within one region. If you have workloads that are located in the same region, you see lower latency and better performance than in a cross-regional setup. The regional endpoint is automatically set by the ibm Helm plug-in that you installed earlier based on the location that your cluster is in. For example, if your cluster is in US South, then your storage classes were configured to use US South as the regional endpoint for your buckets. For more information, see Regions and endpoints.
Review the detailed IBM Cloud Object Storage bucket configuration for a storage class.
```
kubectl describe storageclass <storageclass_name>
```
Example output
```
Name:                  ibmc-s3fs-standard-cross-region
IsDefaultClass:        No
Annotations:           <none>
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=https://s3-api.dal-us-geo.objectstorage.service.networklayer.com,ibm.io/object-store-storage-class=us-standard,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>
```
ibm.io/chunk-size-mb

The size of a data chunk that is read from or written to IBM Cloud Object Storage in megabytes. Storage classes with perf in their name are set up with 52 megabytes. Storage classes without perf in their name use 16 megabyte chunks. For example, if you want to read a file that is 1GB, the plug-in reads this file in multiple 16 or 52-megabyte chunks.

ibm.io/curl-debug

Enable the logging of requests that are sent to the IBM Cloud Object Storage service instance. If enabled, logs are sent to syslog and you can forward the logs to an external logging server. By default, all storage classes are set to false to disable this logging feature.

ibm.io/debug-level

The logging level that is set by the IBM Cloud Object Storage plug-in. All storage classes are set up with the WARN logging level.

ibm.io/iam-endpoint

The API endpoint for IBM Cloud Identity and Access Management (IAM).

ibm.io/kernel-cache

Enable or disable the kernel buffer cache for the volume mount point. If enabled, data that is read from IBM Cloud Object Storage is stored in the kernel cache to ensure fast read access to your data. If disabled, data is not cached and always read from IBM Cloud Object Storage. Kernel cache is enabled for standard and smart storage classes, and disabled for cold and vault storage classes.

ibm.io/multireq-max

The maximum number of parallel requests that can be sent to the IBM Cloud Object Storage service instance to list files in a single directory. All storage classes are set up with a maximum of 20 parallel requests.

ibm.io/object-store-endpoint

The API endpoint to use to access the bucket in your IBM Cloud Object Storage service instance. The endpoint is automatically set based on the region of your cluster. If you want to access an existing bucket that is located in a different region than the one where your cluster is in, you must create a custom storage class and use the API endpoint for your bucket.

ibm.io/object-store-storage-class

The name of the storage class.

ibm.io/parallel-count

The maximum number of parallel requests that can be sent to the IBM Cloud Object Storage service instance for a single read or write operation. Storage classes with perf in their name are set up with a maximum of 20 parallel requests. Storage classes without perf are set up with two parallel requests by default.

ibm.io/s3fs-fuse-retry-count

The maximum number of retries for a read or write operation before the operation is considered unsuccessful. All storage classes are set up with a maximum of five retries.

ibm.io/stat-cache-size

The maximum number of records that are kept in the IBM Cloud Object Storage metadata cache. Every record can take up to 0.5 kilobytes. All storage classes set the maximum number of records to 100000 by default.

ibm.io/tls-cipher-suite

The TLS cipher suite that must be used when a connection to IBM Cloud Object Storage is established via the HTTPS endpoint. The value for the cipher suite must follow the OpenSSL format. If your worker nodes run an Ubuntu operating system, your storage classes are set up to use the AESGCMcipher suite by default. For worker nodes that run a Red Hat operating system, the ecdhe_rsa_aes_128_gcm_sha_256 cipher suite is used by default.

For more information about each storage class, see the storage class reference. If you want to change any of the pre-set values, create your own customized storage class.
Decide on a name for your bucket. The name of a bucket must be unique in IBM Cloud Object Storage. You can also choose to automatically create a name for your bucket by the IBM Cloud Object Storage plug-in. To organize data in a bucket, you can create subdirectories.

The storage class that you chose earlier determines the pricing for the entire bucket. You can't define different storage classes for subdirectories. If you want to store data with different access requirements, consider creating multiple buckets by using multiple PVCs.
Choose if you want to keep your data and the bucket after the cluster or the persistent volume claim (PVC) is deleted. When you delete the PVC, the PV is always deleted. You can choose if you want to also automatically delete the data and the bucket when you delete the PVC. Your IBM Cloud Object Storage service instance is independent from the retention policy that you select for your data and is never removed when you delete a PVC.

Now that you decided on the configuration that you want, you are ready to create a PVC to provision IBM Cloud Object Storage.

Verifying your installation

Review the pod details to verify that the plug-in installation succeeded.

Verify the installation succeeded by listing the driver pods.
```
kubectl get pod --all-namespaces -o wide | grep object
```
Example output
```
ibmcloud-object-storage-driver-9n8g8                              1/1       Running   0          2m
ibmcloud-object-storage-plugin-7c774d484b-pcnnx                   1/1       Running   0          2m
```
The installation is successful when you see one ibmcloud-object-storage-plugin pod and one or more ibmcloud-object-storage-driver pods. 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 ibm-object-s3fs <pod_name> to find the root cause for the failure.

Verify that the storage classes are created successfully.

kubectl get storageclass | grep s3

Example output

ibmc-s3fs-cold-cross-region            ibm.io/ibmc-s3fs   8m
ibmc-s3fs-cold-regional                ibm.io/ibmc-s3fs   8m
ibmc-s3fs-smart-cross-region            ibm.io/ibmc-s3fs   8m
ibmc-s3fs-smart-perf-cross-region       ibm.io/ibmc-s3fs   8m
ibmc-s3fs-smart-perf-regional           ibm.io/ibmc-s3fs   8m
ibmc-s3fs-smart-regional                ibm.io/ibmc-s3fs   8m
ibmc-s3fs-standard-cross-region        ibm.io/ibmc-s3fs   8m
ibmc-s3fs-standard-perf-cross-region   ibm.io/ibmc-s3fs   8m
ibmc-s3fs-standard-perf-regional       ibm.io/ibmc-s3fs   8m
ibmc-s3fs-standard-regional            ibm.io/ibmc-s3fs   8m
ibmc-s3fs-vault-cross-region           ibm.io/ibmc-s3fs   8m
ibmc-s3fs-vault-regional               ibm.io/ibmc-s3fs   8m

If you want to set one of the IBM Cloud Object Storage storage classes as your default storage class, run kubectl patch storageclass <storageclass> -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'. Replace <storageclass> with the name of the IBM Cloud Object Storage storage class.

Follow the instructions to add object storage to your apps.

If you're having trouble installing the IBM Cloud Object Storage plug-in, see Object storage: Installing the Object storage ibmc Helm plug-in fails and Object storage: Installing the IBM Cloud Object Storage plug-in fails.