You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The following table lists the resolved issues for HPE CSI Driver for Kubernetes v2.0.0.
ID
Component
Title
Resolution
BZ-134925
csi.k8s
Disable NodeGetVolumeStats capability
Set the parameter disableNodeGetVolumeStats to true at the time of deployment.
The following table lists the resolved issues for HPE Primera/3PAR CSP for HPE CSI Driver release 2.0.0.
ID
Component
Title
Resolution
BZ-134504
csi.k8s
Suboptimal attach/detach performance when all FC ports are auto-discovered and used for multipathing
For a higher end primera array, there could be several FC ports available for use with multipathing. By default the CSP auto-discovers all available FC ports and uses them for multipathing. This increases the time taken for attach/detach operations significantly. To address this, a new parameter fcPortsList is added in Storageclass. This will be applicable only for newer volumes that are created.
Known Issues
The following table lists the known issues for HPE CSI Driver for Kubernetes v2.0.0. Please note that, Known Issues from previous releases are still applicable with suggested workarounds, if they are not part of Resolved Issues above.
ID
Component
Title
Description
Workaround
CON-2013
csi.k8s
"btrfs" filesystem limitation - it does not use Inodes as other filesystems
Scale setup with mixed workloads may bottleneck the CSP
In a large scale environment when the PVCs and associated pods are created at once, it creates a large number of requests to the CSP.
Create the PVCs prior to creating the workload to lessen the load on the CSP.
CON-2116
csi.k8s
Volume with xfs filesystem created on Ubuntu does not mount on CentOS
xfs Filesystem created on Ubuntu 20.04 fails to mount on another CentOS worker node with error wrong fs type, bad option, bad superblock.
None
CON-2121
csi.k8s
Ephemeral volumes are not deleted after node shutdown
Ephemeral volumes are created as part of NodePublish. On node shutdown if the pods are deleted after MaxWaitForUnmount, the scrubber does not clean up orphan volumes for which there is no associated pod.
None
The following table lists the known issues for HPE Alletra 6000 and Nimble CSP for HPE CSI Driver release 2.0.0
ID
Component
Title
Description
Workaround
CON-1290
csi.k8s
Incorrect secret to access PVC/POD created by different tenant user will leave orphan objects
If incorrect secrets are passed to manage a PVC which was created by a different tenant. With access restrictions, the Nimble CSP will not be able to find the volume created by another tenant. The PVC will be successfully removed on cluster but does not delete the volume on the Storage System
None
CON-1495
csi.k8s
Tenant users have access to volumes created by different user on Nimble GUI
Multi-tenancy feature is only supported through CSI/ and Alletra 600 and Nimble CSP and management of tenant objects through Nimble GUI is not supported.
None
CON-1502
csi.k8s
Multi-tenancy: Volume Attach with CHAP User fails
CSI Driver supports only 1 CHAP User per cluster. Since tenant user has restricted access of CHAP Users, more than 1 tenant with CHAP User is not supported.
None
CON-1573
csi.k8s
importVolume using tenant secret for downstream array
ImportVolume on downstream array does not work with tenant user
ImportVolume using admin user secret
CON-1821
csi.k8s
Not able to create Initiator group for long hostname
initiator group name is limited to 60 characters.
Limit the hostname to lesser than 60 characters
CON-2141
csi.k8s
Pod deletion takes a lot of time with Iscsi Volume Scoped Target
On a scale setup with large number of pods, logout of Volume Scoped Iscsi Targets and device deletion takes a long time.
Use Iscsi Group Scoped Target on a scale setup.
The following table lists the known issues for HPE Alletra 9000 and Primera/3PAR CSP for HPE CSI Driver release 2.0.0