A SaaS platform blade's vHBA logs into the fabric, but the array never matches it. The zone names the blade WWNN, while the vHBA presents a different WWPN at login. What should the zone name instead?
Select an answer to reveal the explanation.
Short Explanation
The WWNN is the server's name, and the WWPN is the name of the port that actually walks in. You zoned the node name, but the fabric login presents the port name, so the array never matches this vHBA. Zone the WWPN the blade logs in with, not the UUID or the Ethernet MAC.
Full Explanation
A service profile has one WWNN for the node and a WWPN for each vHBA. Fabric login registers the WWPN, and zoning and LUN masking match that port name. The SaaS zone contains the WWNN instead, so the name that logged in is not the name in the zone and the array never matches the vHBA. Replacing the zone member with the WWPN the vHBA presents fixes the match. A UUID is a server identity in the service profile. Arrays do not mask LUNs to a UUID. The fabric interconnect cluster virtual IP is the management address of the UCS domain. FLOGI does not use it as a Fibre Channel name. An Ethernet MAC belongs to a vNIC. It is not the Fibre Channel port name, and zoning the MAC does not match the vHBA. Caveat: a blade with two vHBAs has two WWPNs, and the zone has to name the port that is actually in this VSAN, not the port on the other fabric. Operational check: read the WWPN from the logged-in vHBA, put that WWPN in the zone with the array port, activate the zoneset, and confirm the array sees the login.