[BUG]: Auto select protocol makes the node driver to crash #1689
Labels
area/csi-powermax
Issue pertains to the CSI Driver for Dell EMC PowerMax
area/csm-operator
type/bug
Something isn't working. This is the default label associated with a bug issue.
Milestone
Bug Description
After installation of the CSI Driver for PowerMax controller runs ok, but the node drivers loop on:
They do not create the hosts in Unisphere & then crash.
During the first lines of the node log there was was an NVMe error
"Couldn't find any ip interfaces on any of the port-groups"
.NVMe protocol is not accessible on my lab.
During the deployment, I let
X_CSI_TRANSPORT_PROTOCOL: ""
per documentation.After forcing the protocol to
X_CSI_TRANSPORT_PROTOCOL: "FC"
it passed ok.The selection protocol seams to be mandatory. So it looks as a regression.
Idk about the implication of forcing a value in case you have multiple storage protocols (e.g. NFS + FC or NVMe + iSCSI).
Logs
Meaningful error message: https://github.com/dell/csi-powermax/blob/c4d17d7dc72a952651cdcca213ceef7a19c04459/service/node.go#L2004
Screenshots
No response
Additional Environment Information
OCP 4.17 on Baremetal
CSM Operator v1.12
Unisphere 10.1
FC connectivity
Steps to Reproduce
Install the driver and do not set
X_CSI_TRANSPORT_PROTOCOL
; just like in the sample.Expected Behavior
X_CSI_TRANSPORT_PROTOCOL: ""
must be permitted and pick the available protocol for the nodeCSM Driver(s)
PowerMax v2.12
Installation Type
Operator
Container Storage Modules Enabled
No response
Container Orchestrator
Openshift 4.17
Operating System
CoreOS
The text was updated successfully, but these errors were encountered: