rancher-partner-charts/charts/nutanix/nutanix-csi-storage

Nutanix CSI Storage Driver Helm chart

Introduction

The Container Storage Interface (CSI) Volume Driver for Kubernetes leverages Nutanix Volumes and Nutanix Files to provide scalable and persistent storage for stateful applications.

When Files is used for persistent storage, applications on multiple pods can access the same storage, and also have the benefit of multi-pod read and write access.

Important notice

Starting with version 2.5 of this chart we separate the Snapshot components to a second independent Chart. If you plan to update an existing Nutanix CSI Chart version < v2.5.x with this Chart, you need to check below recommendation.

• Once you upgrade to version 2.5+, the snapshot-controler will be removed, but previously installed Snapshot CRD stay in place. You will then need to install the nutanix-csi-snapshot Helm Chart following the Important notice procedure.
• If you create Storageclass automatically with a previous Nutanix CSI Chart version < v2.5.x, take care to remove Storageclass before Helm upgrade.

If you previously installed Nutanix CSI Storage Driver with yaml file please follow the Upgrading from yaml based deployment section below.

If this is your first deployment and your Kubernetes Distribution does not bundle the snapshot components, you need to install first the Nutanix CSI Snapshot Controller Helm chart.

Please note that starting with v2.2.0, Nutanix CSI driver has changed format of driver name from com.nutanix.csi to csi.nutanix.com. All deployment yamls uses this new driver name format. However, if you initially installed CSI driver in version < v2.2.0 then you should need to continue to use old driver name com.nutanix.csi by setting legacy parameter to true. If not existing PVC/PV will not work with the new driver name.

Nutanix CSI driver documentation

https://portal.nutanix.com/page/documents/details?targetId=CSI-Volume-Driver-v2_5:CSI-Volume-Driver-v2_5

Features list

• Nutanix CSI Driver v2.5.0
• Nutanix Volumes support
• Nutanix Files support
• Volume clone
• Volume snapshot and Restore
• IP Address Whitelisting
• LVM Volume supporting multi vdisks volume group
• NFS Dynamic share provisioning
• PV resize support for Volumes and Dynamic Files mode
• iSCSI Auto CHAP Authentication
• OS independence
• Volume metrics and CSI operations metrics support

Prerequisites

• Kubernetes 1.17 or later
• Kubernetes worker nodes must have the iSCSI package installed (Nutanix Volumes mode) and/or NFS tools (Nutanix Files mode)
• This chart have been validated on RHEL/CentOS 7/8 and Ubuntu 18.04/20.04/21.04/21.10, but the new architecture enables easy portability to other distributions.
• This Chart is not made to be installed on the local k3s cluster (by default iscsi prerequisite is missing)

Installing the Chart

To install the chart with the name nutanix-csi:

helm repo add nutanix https://nutanix.github.io/helm/

helm install nutanix-csi nutanix/nutanix-csi-storage -n <namespace of your choice>

Upgrade

Upgrades can be done using the normal Helm upgrade mechanism

helm repo update
helm upgrade nutanix-csi nutanix/nutanix-csi-storage

Upgrading from yaml based deployment

Starting with CSI driver v2.5.0, yaml based deployment is discontinued. So to upgrade from yaml based deployment, you need to patch your existing CSI deployment with helm annotations. Please follow the following procedure.

HELM_CHART_NAME="nutanix-csi"
HELM_CHART_NAMESPACE="ntnx-system"
DRIVER_NAME="csi.nutanix.com"

kubectl delete sts csi-provisioner-ntnx-plugin -n ${HELM_CHART_NAMESPACE}
kubectl patch ds csi-node-ntnx-plugin -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'

kubectl patch csidriver ${DRIVER_NAME} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'

kubectl patch sa csi-provisioner -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'
kubectl patch sa csi-node-ntnx-plugin -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'

kubectl patch clusterrole external-provisioner-runner -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'
kubectl patch clusterrole csi-node-runner -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'

kubectl patch clusterrolebinding csi-provisioner-role -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'
kubectl patch clusterrolebinding csi-node-role -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'

kubectl patch service csi-provisioner-ntnx-plugin -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'

kubectl patch service csi-metrics-service -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}'

kubectl patch servicemonitor csi-driver -n ${HELM_CHART_NAMESPACE} -p '{"metadata": {"annotations":{"meta.helm.sh/release-name":"'"${HELM_CHART_NAME}"'","meta.helm.sh/release-namespace":"'"${HELM_CHART_NAMESPACE}"'"}, "labels":{"app.kubernetes.io/managed-by":"Helm"}}}' --type=merge

Now follow Installing the Chart section to finish upgrading the CSI driver.

Uninstalling the Chart

To uninstall/delete the nutanix-csi deployment:

helm delete nutanix-csi -n <namespace of your choice>

Configuration

The following table lists the configurable parameters of the Nutanix-CSI chart and their default values.

Parameter	Description	Default
legacy	Use old reverse notation for CSI driver name	false
volumeClass	Activate Nutanix Volumes Storage Class	false
volumeClassName	Name of the Nutanix Volumes Storage Class	nutanix-volume
volumeClassDescription	Description prefix for each created VG	volumeClassName
volumeClassRetention	Retention policy for the Volumes Storage Class (Delete, Retain)	Delete
fileClass	Activate Nutanix Files Storage Class	false
fileClassName	Name of the Nutanix Files Storage Class	nutanix-file
fileClassRetention	Retention policy for the Files Storage Class (Delete, Retain)	Delete
dynamicFileClass	Activate Nutanix Dynamic Files Storage Class	false
dynamicFileClassName	Name of the Nutanix Dynamic Files Storage Class	nutanix-dynamicfile
dynamicFileClassDescription	Description prefix for each created Fileshare	dynamicFileClassName
dynamicFileClassRetention	Retention policy for the Dynamic Files Storage Class (Delete, Retain)	Delete
defaultStorageClass	Choose your default Storage Class (none, volume, file, dynfile)	none
prismEndPoint	Cluster Virtual IP Address	10.0.0.1
username	Name used for the admin role (if created)	admin
password	Password for the admin role (if created)	nutanix/4u
secretName	Name of the secret to use for admin role	ntnx-secret
createSecret	Create secret for admin role (if false use existing)	true
storageContainer	Nutanix storage container name	default
fsType	Type of file system you are using (ext4, xfs)	xfs
networkSegmentation	Activate Volumes Network Segmentation support	false
lvmVolume	Activate LVM to use multiple vdisks by Volume	false
lvmDisks	Number of vdisks by volume if lvm enabled	4
fileHost	NFS server IP address	10.0.0.3
filePath	Path of the NFS share	share
fileServerName	Name of the Nutanix FIle Server	file
kubeletDir	allows overriding the host location of kubelet's internal state	/var/lib/kubelet
nodeSelector	Add nodeSelector to all pods	{}
tolerations	Add tolerations to all pods	[]
imagePullPolicy	Specify imagePullPolicy for all pods	IfNotPresent
controller.replicas	Number of Controllers replicas to deploy.	2
controller.nodeSelector	Add nodeSelector to provisioner pod	{}
controller.tolerations	Add tolerations to provisioner pod	[]
node.nodeSelector	Add nodeSelector to node pods	{}
node.tolerations	Add tolerations to node pods	[]
servicemonitor.enabled	Create ServiceMonitor to scrape CSI metrics	false
servicemonitor.labels	Labels to add to the ServiceMonitor (for match the Prometheus serviceMonitorSelector logic)	k8s-app: csi-driver

Specify each parameter using the --set key=value[,key=value] argument to helm install or provide a a file whit -f value.yaml.

Configuration examples:

Install the driver in the ntnx-system namespace:

helm install nutanix-storage nutanix/nutanix-csi-storage -n ntnx-system --create-namespace

Install the driver in the ntnx-system namespace and create a volume storageclass:

helm install nutanix-storage nutanix/nutanix-csi-storage -n ntnx-system --create-namespace --set volumeClass=true --set prismEndPoint=X.X.X.X --set username=admin --set password=xxxxxxxxx --set storageContainer=container_name --set fsType=xfs

In the above command prismEndpoint refers to the Prism Element cluster virtual ip address where storage will be consumed.

Install the driver in the ntnx-system namespace, create a volume and a dynamic file storageclass and set the volume storage class as default:

helm install nutanix-storage nutanix/nutanix-csi-storage -n ntnx-system --create-namespace --set volumeClass=true --set prismEndPoint=X.X.X.X --set username=admin --set password=xxxxxxxxx --set storageContainer=container_name --set fsType=xfs --set defaultStorageClass=volume --set dynamicFileClass=true --set fileServerName=name_of_the_file_server

In the above command prismEndpoint refers to the Prism Element cluster virtual ip address where storage will be consumed.

All the options can also be specified in a value.yaml file:

helm install nutanix-storage nutanix/nutanix-csi-storage -n ntnx-system --create-namespace -f value.yaml

Support

The Nutanix CSI Volume Driver is fully supported by Nutanix. Please use the standard support procedure to file a ticket here.

Community

Please file any issues, questions or feature requests you may have here for the Nutanix CSI Driver or here for the Helm chart.

Contributing

We value all feedback and contributions. If you find any issues or want to contribute, please feel free to open an issue or file a PR.