Release notes
Release notes for Dell Container Storage Modules Operator
Release Notes - Container Storage Modules Operator v1.4.4
New Features/Changes
- #947 - [FEATURE]: Support for Kubernetes 1.28
- #1066 - [FEATURE]: Support for Openshift 4.14
- #996 - [FEATURE]: Dell CSI to Dell CSM Operator Migration Process
- #1062 - [FEATURE]: CSM PowerMax: Support PowerMax v10.1
- #1158 - [FEATURE]: Automatically create certificates with CSM Operator Observability deployment
Fixed Issues
- #975 - [BUG]: Not able to take volumesnapshots
- #982 - [BUG]: Update resources limits for controller-manager to fix OOMKilled error
- #988 - [BUG]: CSM Operator fails to install CSM Replication on the remote cluster
- #989 - [BUG]: Allow volume prefix to be set via CSM operator
- #990 - [BUG]: X_CSI_AUTH_TYPE cannot be set in CSM Operator
- #1110 - [BUG]: Multi Controller defect - sidecars timeout
- #1117 - [BUG]: Operator crashes when deployed from OpenShift with OLM
- #1120 - [BUG]: Skip Certificate Validation is not propagated to Authorization module in CSM Operator
- #1122 - [BUG]: CSM Operator does not calculate status correctly when module is deployed with driver
- #1103 - [BUG]: CSM Operator doesn’t apply fSGroupPolicy value to CSIDriver Object
- #1133 - [BUG]: CSM Operator does not calculate status correctly when application-mobility is deployed by itself)
- #1137 - [BUG]: CSM Operator intermittently does not calculate status correctly when deploying a driver)
- #1143 - [BUG]: CSM Operator does not calculate status correctly when deploying the authorization proxy server)
- #1146 - [BUG]: CSM Operator does not calculate status correctly when deploying observability with csi-powerscale)
- #1147 - [BUG]: CSM Operator labels csm objects with CSMVersion 1.8.0, an old version)
- #1156 - [BUG]: CSM object in success state when all CSI Powerflex pods are failing due to bad secret credentials)
- #1157 - [BUG]: If Authorization Proxy Server is installed in an alternate namespace by CSM Operator, the deployment fails)
- #1159 - [BUG]: CSM status is not always accurate when Observability installed with only one or two components)
- #1152 - [BUG]: CSI driver changes to facilitate SDC brownfield deployments)
- #1171 - [BUG]: CSM object occasionally stays in failed state when app-mobility is successfully deployed with csm-operator)
- #1189 - [BUG]: Change the Apex Connectivity Client access to the kube-proxy port to only connections within the client pod)
- #1190 - [BUG]: Change Apex Connectivity Client access to secrets to only the secrets it needs to manage)
Known Issues
Issue | Workaround |
---|---|
The status field of a csm object as deployed by CSM Operator may, in limited cases, display an incorrect status for a deployment. | As a workaround, the health of the deployment can be determined by checking the health of the pods. |
When CSM Operator creates a deployment that includes secrets (e.g., application-mobility, observability, cert-manager, velero), these secrets are not deleted on uninstall and will be left behind. For example, the karavi-topology-tls , otel-collector-tls , and cert-manager-webhook-ca secrets will not be deleted. |
This should not cause any issues on the system, but all secrets present on the cluster can be found with kubectl get secrets -A , and any unwanted secrets can be deleted with kubectl delete secret -n <secret-namespace> <secret-name> |
CSM Operator does not support dynamic namespaces for Authorization. Despite successful installation in a namespace other than “authorization”, errors may arise during volume creation. | Use the default namespace “authorization” for installing Authorization using CSM Operator |
The images of sideCars are currently missing in the sample YAMLs in the offline bundle. As a consequence, the csm-operator is pulling them from registry.k8s.io. | We recommend manually updating the images of sideCars in the sample YAML file, for example, storage_csm_powerflex_v291.yaml , before proceeding with the driver installation. Here is an example snippet for the sideCars section in the YAML file: |
sideCars:
# 'k8s' represents a string prepended to each volume created by the CSI driver
- name: provisioner
image: <localregistry>/csi-provisioner:v3.6.2
args: ["--volume-name-prefix=k8s"]
- name: attacher
image: <localregistry>/csi-attacher:v4.4.2
- name: registrar
image: <localregistry>/csi-node-driver-registrar:v2.9.1
- name: resizer
image: <localregistry>/csi-resizer:v1.9.2
- name: snapshotter
image: <localregistry>/csi-snapshotter:v6.3.2
# sdc-monitor is disabled by default, due to high CPU usage
- name: sdc-monitor
enabled: false
image: <localregistry>/sdc:4.5
envs:
- name: HOST_PID
value: "1"
- name: MDM
value: "10.xx.xx.xx,10.xx.xx.xx" # Do not add mdm value here if it is present in secret
# health monitor is disabled by default, refer to driver documentation before enabling it
# Also set the env variable controller.envs.X_CSI_HEALTH_MONITOR_ENABLED to "true".
- name: csi-external-health-monitor-controller
enabled: false
image: <localregistry>/csi-external-health-monitor-controller:v0.10.0
args: ["--monitor-interval=60s"]