Releases: dell/csi-powerflex
Releases · dell/csi-powerflex
Release v2.6.0
Release Notes - CSI PowerFlex v2.6.0
New Features/Changes
- PowerFlex pre-approved GUIDs support added.
- Rename SDC support added.
- K8 1.26 support added.
- RKE 1.4.1 support added.
- MKE 3.6.0 support added.
Fixed Issues
Known Issues
| Issue | Workaround |
|---|---|
| Delete namespace that has PVCs and pods created with the driver. The External health monitor sidecar crashes as a result of this operation. | Deleting the namespace deletes the PVCs first and then removes the pods in the namespace. This brings a condition where pods exist without their PVCs and causes the external-health-monitor sidecar to crash. This is a known issue and has been reported at kubernetes-csi/external-health-monitor#100 |
| When a node goes down, the block volumes attached to the node cannot be attached to another node | This is a known issue and has been reported at kubernetes-csi/external-attacher#215. Workaround: 1. Force delete the pod running on the node that went down 2. Delete the volumeattachment to the node that went down. Now the volume can be attached to the new node. |
| sdc:3.6.0.6 is causing issues while installing the csi-powerflex driver on ubuntu,RHEL8.3 | Workaround: Change the powerflexSdc to sdc:3.6 in values.yaml https://github.com/dell/csi-powerflex/blob/72b27acee7553006cc09df97f85405f58478d2e4/helm/csi-vxflexos/values.yaml#L13 |
Note:
- Support for Kubernetes alpha features like Volume Health Monitoring and RWOP (ReadWriteOncePod) access mode will not be available in Openshift environment as Openshift doesn't support enabling of alpha features for Production Grade clusters.
Documentation
Release Notes
General Documentation
Change Log
See the CHANGELOG for more details.
Release v2.5.0
Release Notes - CSI PowerFlex v2.5.0
New Features/Changes
- Read Only Block support
- Added support for setting QoS limits by CSI-PowerFLex driver
- Added support for standardizing helm installation for CSI-PowerFlex driver
- Automated SDC deployment on RHEL 7.9 and 8.x
- SLES 15 SP4 support added
- OCP 4.11 support added
- K8 1.25 support added
- Added support for PowerFlex storage system v4.0
Fixed Issues
Known Issues
| Issue | Workaround |
|---|---|
| Delete namespace that has PVCs and pods created with the driver. The External health monitor sidecar crashes as a result of this operation. | Deleting the namespace deletes the PVCs first and then removes the pods in the namespace. This brings a condition where pods exist without their PVCs and causes the external-health-monitor sidecar to crash. This is a known issue and has been reported at kubernetes-csi/external-health-monitor#100 |
| When a node goes down, the block volumes attached to the node cannot be attached to another node | This is a known issue and has been reported at kubernetes-csi/external-attacher#215. Workaround: 1. Force delete the pod running on the node that went down 2. Delete the volumeattachment to the node that went down. Now the volume can be attached to the new node. |
Note:
- Support for Kubernetes alpha features like Volume Health Monitoring and RWOP (ReadWriteOncePod) access mode will not be available in OpenShift environment as OpenShift doesn't support enabling of alpha features for Production Grade clusters.
Documentation
Release Notes
General Documentation
Change Log
See the CHANGELOG for more details.
Release v2.4.0
Release Notes - CSI PowerFlex v2.4.0
New Features/Changes
- Added optional parameter protectionDomain to storageclass
- Added InstallationID annotation for volume attributes.
- RHEL 8.6 support added
Fixed Issues
Known Issues
| Issue | Workaround |
|---|---|
| Delete namespace that has PVCs and pods created with the driver. The External health monitor sidecar crashes as a result of this operation. | Deleting the namespace deletes the PVCs first and then removes the pods in the namespace. This brings a condition where pods exist without their PVCs and causes the external-health-monitor sidecar to crash. This is a known issue and has been reported at kubernetes-csi/external-health-monitor#100 |
| When a node goes down, the block volumes attached to the node cannot be attached to another node | This is a known issue and has been reported at kubernetes-csi/external-attacher#215. Workaround: 1. Force delete the pod running on the node that went down 2. Delete the volumeattachment to the node that went down. Now the volume can be attached to the new node. |
Note:
- Support for Kubernetes alpha features like Volume Health Monitoring and RWOP (ReadWriteOncePod) access mode will not be available in Openshift environment as Openshift doesn't support enabling of alpha features for Production Grade clusters.
Documentation
Release Notes
General Documentation
Change Log
See the CHANGELOG for more details.
Release v2.3.0
Release Notes - CSI PowerFlex v2.3.0
New Features/Changes
- Added support to configure fsGroupPolicy
- Removed beta volumesnapshotclass sample files.
- Added support for Kubernetes 1.24.
- Added support for OpenShift 4.10.
- Fixed handling of idempotent snapshots.
Fixed Issues
- Added label to driver node pod for Resiliency protection.
- Updated values file to use patched image of vg-snapshotter.
Known Issues
| Issue | Workaround |
|---|---|
| Delete namespace that has PVCs and pods created with the driver. The External health monitor sidecar crashes as a result of this operation. | Deleting the namespace deletes the PVCs first and then removes the pods in the namespace. This brings a condition where pods exist without their PVCs and causes the external-health-monitor sidecar to crash. This is a known issue and has been reported at kubernetes-csi/external-health-monitor#100 |
| When a node goes down, the block volumes attached to the node cannot be attached to another node | This is a known issue and has been reported at kubernetes-csi/external-attacher#215. Workaround: 1. Force delete the pod running on the node that went down 2. Delete the volumeattachment to the node that went down. Now the volume can be attached to the new node. |
Note:
- Support for Kubernetes alpha features like Volume Health Monitoring and RWOP (ReadWriteOncePod) access mode will not be available in Openshift environment as Openshift doesn't support enabling of alpha features for Production Grade clusters.
Change Log
See the CHANGELOG for more details.
Release v2.2.0
Release Notes - CSI PowerFlex v2.2.0
New Features/Changes
- Added support for Kubernetes 1.23.
- Added support for Amazon Elastic Kubernetes Service Anywhere.
Fixed Issues
There are no fixed issues in this release.
Known Issues
| Issue | Workaround |
|---|---|
| Delete namespace that has PVCs and pods created with the driver. The External health monitor sidecar crashes as a result of this operation. | Deleting the namespace deletes the PVCs first and then removes the pods in the namespace. This brings a condition where pods exist without their PVCs and causes the external-health-monitor sidecar to crash. This is a known issue and has been reported at kubernetes-csi/external-health-monitor#100 |
Note:
- Support for Kubernetes alpha features like Volume Health Monitoring and RWOP (ReadWriteOncePod) access mode will not be available in Openshift environment as Openshift doesn't support enabling of alpha features for Production Grade clusters.
Change Log
See the CHANGELOG for more details.
Release v2.1.0
New Features/Changes
- Added support for OpenShift v4.9.
- Added support for CSI spec 1.5.
- Added support for new access modes in CSI Spec 1.5.
- Added support for PV/PVC metrics.
- Added support for CSM Authorization sidecar via Helm.
- Added v1 extensions to vg snapshot from v1alpha2.
- Added support to update helm charts to do a helm install without shell scripts.
- Added support for volume health monitoring.
Fixed Issues
There are no fixed issues in this release.
Documentation
Release v2.0.0
New Features/Changes
- Added support for Kubernetes v1.22.
- Added support for OpenShift v4.8.
- Added support for dynamic multi-array configuration.
- Added certificate validation for PowerFlex Gateway REST API calls.
- Aligned configuration parameters with those in other CSI drivers in effort to have consistent config experience across drivers.
- Added the ability to configure kubelet directory path.
- Added the ability to enable/disable installation of resizer sidecar with driver installation.
- Added the ability to enable/disable installation of snapshotter sidecar with driver installation.
- Added the ability to use system names instead of system IDs in config files.
Release v1.5.1
Changes:
This patch addresses issues with specifying a PowerFlex name instead of ID when configuring the driver that can lead to issues when unpublishing volumes.
Instructions to use:
- If PowerFlex system name is used in secret and in storage class and the name has changed, add a new field to array secret "allSystemNames" with a comma separated list of that systems previous names, for example:
password: "xxxxxx" systemID: "pflex_new_system" allSystemNames: "pflex-previous1-name,pflex-previous2-name" endpoint: "https://1.2.3.4" skipCertificateValidation: true mdm: "1.2.4.5" isDefault: true
Note: allSystemNames: "pflex-old-system1" , has the other names with which volumes were provisioned using this csi-driver.
restart the vxflexos driver pods after secret has been updated.
Helm:
New driver image is already supplied in helm/csi-vxflexos/driver-image.yaml. To upgrade your driver to the patched image, run:
./csi-install.sh --namespace vxflexos --values <your values file> --upgrade
Operator:
Replace image dellemc/csi-vxflexos:v1.5.0 with dellemc/csi-vxflexos:v1.5.1 in sample yaml.
Release v1.5.0
New Features/Changes
- Support for RedHat Enterprise Linux 8.4 Host OS
- Support for Mirantis Kubernetes Engine 3.4 (Formerly Docker EE)
- Support for OpenShift 4.6 EUS release
- Support for Rancher RKE v1.2.8
- Support for CSI Specification 1.3
- CSM Volume Group Snapshotter (tech preview)
- Support for mkfs format option
- Support for dynamic log config: log level and log format
- Config.yaml support
Release v1.4.0
New Features/Changes
- Added support for Kubernetes v1.20
- Added support for OpenShift 4.7 with RHEL and CoreOS worker nodes
- Added support for Red Hat Enterprise Linux (RHEL) 8.3
- Added support for Fedora CoreOS kernel version 5.10
- Added SDC deployment on Fedora CoreOS nodes
- Added support for Ephemeral Inline Volume
- Added support multi-mount volumes
- Added support for managing multiple PowerFlex arrays from one driver
- Removed storage classes from helm template