Skip to main content
Version: 26.1

Frequently Asked Questions

This page answers common questions about PX-CSI, including feature support, resource configuration, and changes introduced in recent releases.

Is the pxctl CLI still supported?​

The pxctl CLI was removed in PX-CSI 25.8.0.

  • PX-CSI 25.8.1 and later: Use the PX CLI instead. The PX CLI uses Kubernetes credentials from your kubeconfig (~/.kube/config) to interact with the PX-CSI driver deployed in your cluster. It communicates with the CSI controller and node plugins to manage StorageClass, VolumeSnapshot, and other CSI resources, as well as CSI driver configuration. For more information, see PX CLI reference.
  • PX-CSI 25.8.0: Use Kubernetes-native resources like PersistentVolumeClaim, PersistentVolume, VolumeSnapshotClass, and VolumeSnapshot instead.

Where does PX-CSI store volume metadata?​

PX-CSI stores volume metadata in Kubernetes CustomResourceDefinitions (CRDs) in the portworx namespace:

  • PureVolume – volume metadata
  • PureSnapshot – snapshot metadata

During an upgrade from 25.6.0 or earlier to 25.8.0 or later, PX-CSI automatically migrates metadata from KVDB to CRDs.

You can use the following commands to inspect the resources:

kubectl get purevolume -n portworx
kubectl get puresnapshot -n portworx

See PureVolume CRD reference and PureSnapshot CRD reference for more information.

How can I control resource limits, node placement, and tolerations?​

Use the ComponentK8sConfig custom resource.

PX-CSI automatically creates a px-pure-csi-k8s-config Custom Resource. Configure resource limits, node selectors, node affinities, and tolerations in this custom resource instead of legacy StorageCluster configurations.

For more information, see Configure resource limits.

Does PX-CSI support FlashArray ActiveCluster?​

PX-CSI supports FlashArray ActiveCluster starting in version 26.1.0.

For more information, see Use FlashArray ActiveCluster for high availability.

How do I copy a volume into a stretched pod for ActiveCluster?​

You can copy an existing volume into a stretched pod (a pod configured for FlashArray ActiveCluster) using one of the following methods:

Stretch-Copy​

  1. Place the source volume into a local pod (for example, temppod).
  2. Stretch temppod between the same arrays as the destination pod.
  3. Use purevol copy to copy the volume.
  4. Remove temppod from the stretched configuration.

Copy-Stretch-Move​

  1. Copy the volume into a temporary pod (for example, temppod) on the array.
  2. Stretch temppod between the same arrays as the destination pod.
  3. Use purevol move to move the copied volume from temppod to the destination pod.
  4. Remove temppod from the stretched configuration.

After copying the volume using either method, you can import it into Kubernetes using static provisioning. For more information, see Static provisioning of FlashArray block volumes.

Are proxy volumes supported?​

No. Proxy volumes are not supported starting in PX-CSI 25.8.0.

Before you upgrade, contact Portworx Support for assistance with migration or refactoring workflows that depend on proxy volumes.

How do I configure multiple iSCSI interfaces?​

Add each interface to the iSCSI database, and set the PURE_ISCSI_ALLOWED_IFACES environment variable as a comma-separated list of interfaces. For more information, see Install Portworx CSI.

Does FBDA support an external backup provider?​

No. FBDA does not support external backup providers.

How are changes to pure.json reflected in the Kubernetes cluster?​

Restart both the CSI node plugin and CSI controller plugin pods to apply the changes.