We have Kubernetes worker nodes on Hyper-V with virtual fibre channel, multipathd is configured as described by the documentation and udev as well.
OS: Ubuntu 24.04
Kernel: 6.8.0-90
K8s version: v1.32.9
px-pure-csi-driver:25.8.1
px-operator:25.5.0
PX-CSI is running fine, NFS is working. We can also see all connectivity towards our PureStorage FlashArray from the fibre channel connection and on the OS it states that the connection is online. The LUN is being created and mapped to the correct server. We can see the LUN with multipath -ll:
dm-5 PURE,FlashArray
size=5.0G features=‘0’ hwhandler=‘1 alua’ wp=rw
`-± policy=‘service-time 0’ prio=50 status=active
|- 4:0:1:1 sde 8:64 active ready running
|- 5:0:1:1 sdg 8:96 active ready running
|- 4:0:0:1 sdd 8:48 active ready running
`- 5:0:0:1 sdf 8:80 active ready running
Issue is that Hyper-V somehow does not provide the remote WWN to the guest OS:
/sys/class/fc_remote_ports/rport-4:0-0
port_name=0x0
node_name=0x0
port_id=0x000000
port_state=Online
roles=FCP Dummy Initiator
/sys/class/fc_remote_ports/rport-5:0-0
port_name=0x0
node_name=0x0
port_id=0x000000
port_state=Online
roles=FCP Dummy Initiator
This causes PX-CSI to fail when the pod tries to attach the volume:
error attaching device: Unable to verify FC connectivity to targets, no remote wwpns matched target wwns
This tells me that PX-CSI cannot find the remote WWPNs to match the target WWNs with, possibly this is a check to see if the correct host is being mapped on the FlashArray?
Is there a possibility to run PX-CSI over FC on Hyper-V or is this impossible?