Mounting volume fails on FCoE envi, due to the protocol type FC

[c-raghav]

Observation:

Mounting volume fails on FCoE environment,due to the protocol type - FCOE not being available

Setup Details:

The setup does have two arrays attached, one with FC with one of the Merlin array (15.213.71.140)and
this setup was being used for google cloud POC, and using the same setup with FCoE with slight modification of
hpe.conf files, where we have modified/added for FCoE setup for new array( 9K - 250)
3pardcv.log

The WWN - 1000E0071BCEDA72 is FC initiator and in zone with Merlin array (15.213.71.140), which currently not using.

Right now we are concentrating on initiator WWN - 1000f40343a6d161, which is FCoE and is in zone with 9K array 15.213.64.250

CB2506_94502N_250 cli% showport -fcoe
N:S:P ENode_MAC_Address PFC_Mask
0:4:1 00-02-AC-08-12-27 0x08
0:4:2 00-02-AC-09-12-27 0x00

Zone details

MSDU_LS_E_CB2312_A:FID100:xpparaghavendra> zoneshow server_array_140
zone: server_array_140
10:00:e0:07:1b:ce:da:72; 21:31:00:02:ac:07:e2:01

Though we are able to create the volume, mount is failing over there, we can ignore the FC WWN as of now.

[root@cssosecostor-b10 ~]# ssh -l topo 15.213.64.250
[email protected]'s password:
CB2506_94502N_250 cli% showhost cssosecostor-b10
Id Name Persona -WWN/iSCSI_Name- Port
46 cssosecostor-b10 Generic-ALUA 1000E0071BCEDA72 ---
1000F40343A6D161 0:4:1

2 total
[root@cssosecostor-b10 ~]# systool -c fc_host -v | grep port_name
port_name = "0x1000e0071bceda72"
port_name = "0x1000e0071bceda73"
port_name = "0x1000f40343a6d161"
[root@cssosecostor-b10 ~]# systool -c fc_host -v | grep -A 5 -B 4 port_name
maxframe_size = "2048 bytes"
node_name = "0x2000e0071bceda72"
npiv_vports_inuse = "0"
port_id = "0x0b0c07"
port_name = "0x1000e0071bceda72"
port_state = "Online"
port_type = "NPort (fabric via point-to-point)"
speed = "8 Gbit"
supported_classes = "Class 3"
supported_fc4s = "0x00 0x00 0x01 0x00 0x00 0x00 0x00 0x01 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 "

maxframe_size       = "2048 bytes"
node_name           = "0x2000e0071bceda73"
npiv_vports_inuse   = "0"
port_id             = "0x000000"
port_name           = "0x1000e0071bceda73"
port_state          = "Linkdown"
port_type           = "Unknown"
speed               = "unknown"
supported_classes   = "Class 3"
supported_fc4s      = "0x00 0x00 0x01 0x00 0x00 0x00 0x00 0x01 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 "

--
maxframe_size = "2048 bytes"
node_name = "0x2000f40343a6d161"
npiv_vports_inuse = "0"
port_id = "0xc80002"
port_name = "0x1000f40343a6d161"
port_state = "Online"
port_type = "NPort (fabric via point-to-point)"
speed = "20 Gbit"
supported_classes = "Class 3"
supported_fc4s = "0x00 0x00 0x01 0x00 0x00 0x00 0x00 0x01 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 "

[wdurairaj]

Right now, the FC Driver reads port of transport type FC only , that seems to the be cause after the initial analysis

https://github.com/hpe-storage/python-hpedockerplugin/blob/plugin_v2/hpedockerplugin/hpe/hpe_3par_common.py#L387

[wdurairaj]

Change required is

 def get_active_fc_target_ports(self):
        ports = self.get_active_target_ports()
        fc_ports = []
        for port in ports:
            if (port['protocol'] == self.client.PORT_PROTO_FC or
               port['protocol'] == self.client.PORT_PROTO_FCOE):
                fc_ports.append(port)

        return fc_ports