Overview of standby mode

How target ports provide access to LUNs

Port A on each target HBA operates as the active port, and Port B operates as a standby port. When the cluster is in normal operation, Port A provides access to local LUNs, and Port B is not available to the initiator. When one system in the cluster fails, Port B on the partner system becomes active and provides access to the LUNs on the failed system. The Port B assumes the WWPN of the Port A on the failed partner. The ports log in to the fabric in point-to-point mode.

Some operating systems, such as HP-UX and AIX, do not support standby mode. For detailed information, see the documentation for your Host Attach Kit.

Limitations of standby mode

The standby cfmode setting is supported on all FCP-licensed systems except for the FAS270c.

Only Solaris and Windows igroup types can access a system in standby mode. The HP-UX, AIX, and default igroup types are not supported.

This setting also requires more switch ports because Port A and Port B on each HBA must connect to the switch, even though B becomes active only in the event of a takeover.

What happens during a failover

The following example shows a configuration in which Port B operates as a standby port. The storage system cluster pair shows each storage system with two target HBAs, in slots M and N. For each storage system, the slot-M HBA connects to Switch 1 and slot-N HBA connects to Switch 2. The solid lines indicate active connections. The dotted lines indicate standby connections.

If Filer X fails, then Filer Y takes over and the following occurs:

Slot-M-port b on Filer Y takes over for slot-M-port a in Filer X
Slot-N-port b on Filer Y takes over for slot-N-port a in Filer X

Port B on each HBA in Filer Y becomes active and enables the host to access the storage until Filer X is repaired and running. Each B port assumes the WWNN and WWPN of the corresponding A port on the failed storage system.

How Data ONTAP displays information about target ports

When the FCP cfmode setting is standby, the local WWNN and WWPN have a pattern of 50:a9:80:nn:nn:nn:nn:nn or 50:0a:09:nn:nn:nn:nn:nn. Each port has a unique WWPN. The standby WWNN and WWPN have a pattern of 20:01:00:nn:nn:nn:nn:nn.

The following fcp config output shows target port information for a storage system in standby mode. The target HBAs are installed in slots 9 and 11. Port 1 in slot 9 is displayed as 9a. Port 2 in slot 9 is displayed as 9b.

filer> fcp config

9a:   ONLINE <ADAPTER UP>  PTP  Fabric

        host address 021b00

        portname 50:a9:80:01:03:00:e0:73  nodename 50:a9:80:00:03:00:e0:73

        mediatype ptp  partner adapter None

9b:   ONLINE <ADAPTER UP>  PTP  Fabric  Standby

        host address 021a00

        portname 20:01:00:e0:8b:28:71:54  nodename 20:01:00:e0:8b:28:71:54

        mediatype ptp  partner adapter 9a

11a:   ONLINE <ADAPTER UP>  PTP  Fabric

        host address 021500

        portname 50:a9:80:03:03:00:e0:73  nodename 50:a9:80:00:03:00:e0:73

        mediatype ptp  partner adapter None

11b:   ONLINE <ADAPTER UP>  PTP  Fabric  Standby

        host address 021600

        portname 20:01:00:e0:8b:28:70:54  nodename 20:01:00:e0:8b:28:70:54

        mediatype ptp  partner adapter 11a