NetApp storage appliances are hardware and software-based storage systems, such as filers, that serve data using protocols for both SAN and NAS environments. Storage appliances include the main unit, or chassis, disk shelves and disk drives, and cables. Storage appliances are installed in telco racks or in NetApp system cabinets.
With Data ONTAP 6.4, you can configure a filer as a target device in a SAN environment using FCP or iSCSI to communicate with one or more hosts running operating systems such as Solaris or Windows. You can also configure LUNs for multiprotocol access. You configure the filer's storage as logical units of storage (LUNs) for block access or as files for file access, or both.
Data ONTAP 6.4 supports the following SAN protocols:
Data ONTAP 6.4 supports the following NAS protocols:
In a NetApp SAN, filers (targets) act on requests from hosts (initiators), or application servers, and process those requests by writing data to or retrieving data from disks in disk shelves. Data ONTAP provides a complete set of storage management tools through its command-line interface and through the FilerView® interface and DataFabricTM Manager interface.
Data ONTAP 6.4 supports the following configurations for connecting targets (filers) to initiators (hosts) in the FCP SAN:
A single-path host has only one initiator Host Bus Adapter (HBA).
A multipath host has two initiator HBAs and requires additional multipathing software, such as the DMP (Dynamic Multipathing) module of VERITAS® volume Manager 3.2 for Solaris. This configuration does not support access of data on a LUN as an NFS file if it is part of a volume stored under VERITAS Volume Manager. Data on a LUN that has a VxFS file system on it is also not accessible through NFS.
Data on LUNs that are not part of a VERITAS volume and do not have a VERITAS file system on them can still be accessed through NFS if the partition table conforms to the rules defined in the Host Bus Adapter Installation and Setup Guide 1.1 for Fibre Channel Protocol on Solaris.
Clustered filers have two target HBAs installed in the PCI slots of the chassis. Stand-alone filers in the direct-attached configuration have one target HBA. Slot assignments vary, depending on the filer model. See the System Configuration Guide at http://now.netapp.com/NOW/ knowledge/docs.shtml for current slot configuration information.
For more information about configurations, see one of the following guides, depending on your environment:
For the latest information about all supported configurations, see the SAN Support Matrix at the NOW site at http://now.netapp.com/NOW/knowledge/ docs.shtml.
Data ONTAP 6.4 supports the following configurations for initiators and targets in the iSCSI SAN:
For FCP, the NetApp SAN network provides basic Fibre Channel SAN operation; however, Network Appliance highly recommends you use a TCP/IP network for connectivity to perform the following tasks:
For iSCSI, use a TCP/IP network for all connectivity.
In a NetApp SAN network using FCP, nodes are connected to each other with Network Appliance-supplied optical cables to communicate with each other. Nodes are one of the following Fibre Channel devices: filers, hosts, switches.
In a NetApp SAN network using iSCSI, nodes are connected to each other with Ethernet cable. iSCSI nodes are filers or hosts.
In either environment, every node must have a unique identifier.
For FCP, every node has a unique World Wide Node Name (WWNN).
NetApp assigns a WWNN to a filer based on the serial number of its NVRAM card. The WWNN is stored on disk.
Each Fibre Channel device has one or more ports that are used to connect it to the SAN network. Each port has a unique World Wide Port Name (WWPN).
Both the WWNN and the WWPN are a 64-bit address represented in the following format: nn:nn:nn:nn:nn:nn:nn:nn, where n represents a hexadecimal value.
For both filers and hosts, each port of an HBA has a unique WWPN.
For filers, Data ONTAP displays the WWNN as FC nodename or nodename, and displays the WWPN as FC portname or portname when using commands such as storage show adapter, fcp show adapter, fcp config, or sysconfig -v.
Each target HBA has two ports, labeled Port 1 and Port 2. Data ONTAP displays information about Port 1 by referring to it as a of the slot number the HBA is installed in and information about Port 2 as slot number b. For example, if the target HBA is installed in slot 5, Port 1 is displayed as 5a and Port 2 is displayed as 5b.
For filers, when the port is active, the addresses of the WWNN and WWPN have a pattern of 50:a9:80:nn:nn:nn:nn. The FC node name for both active ports (5a and 6a) are identical because they are the WWNN for the filer. The FC port names for the active ports are unique. The FC node name and port name for standby ports are based on the adapter's native node and port names so that when adapters log in with the fabric, they have a different node name and port name from the filer. This prevents initiators from seeing the standby adapter as a second path to the target's active node. After takeover occurs, the standby filer and its adapters become active, the node name and port name of its adapter's ports are changed to the node name and port name of the failed partner, and the standby filer exports its partner's LUNs.
When the port is in standby mode, the addresses of the WWNN and WWPN have a pattern of 20:01:00:nn:nn:nn:nn:nn.
For example, assume that Filer_A and Filer_B are clustered and their target HBAs are installed in slots 5 and 6 in the chassis of both filers. Also assume that the ports are configured as follows:
This configuration is illustrated in the following figure.
The following example shows the patterns of the port names (WWPNs) and node names (WWNNs) for the active and standby (passive) ports of Filer_A.
Filer_A Active ports:
5a: FC nodename 50:a9:80:00:02:00:bf:0e
FC portname 50:a9:80:01:02:00:bf:0e
6a: FC nodename 50:a9:80:00:02:00:bf:0e
FC portname 50:a9:80:03:02:00:bf:0e
Filer_A Standby ports:
5b: FC nodename 20:01:00:e0:8b:26:3a:b7
FC portname 20:01:00:e0:8b:26:3a:b7
6b: FC nodename 20:01:00:e0:8b:27:c0:6c
FC portname 20:01:00:e0:8b:27:c0:6c
When the filers are not in takeover mode, you can use the WWNN and WWPN of the standby ports to verify connectivity.
For hosts, Data ONTAP displays the WWPN as portname for commands such as fcp show initiator. The host WWPN is derived by prepending 10:00 to its IEEE address and has a pattern of 10:00:00:00:nn:nn:nn:nn.
When a Fibre Channel device is connected to the SAN network, it registers each of its ports with the switch's Fabric Name Server directory service using its WWNN and WWPN.
The WWNN of the filer is used to create persistent bindings between Solaris hosts and the LUNs on a particular filer. See the Host Bus Adapter Installation and Setup Guide 1.0 for Fibre Channel Protocol on Solaris for information about the filer's WWNN and target IDs.
Note
The use of persistent bindings is not necessary on Windows hosts.
WWPNs of the host are used when creating initiator groups. You use initiator groups to map a LUN to a host, thereby making the LUN accessible to the host.
Fibre Channel switches have one WWNN for the device itself, and one WWPN for each of its ports. For example, the following diagram shows how the WWPNs are assigned to each of the ports on a 16-port switch.
Port 0, WWPN 20:00:00:60:69:51:06:b4
Port 1, WWPN 20:01:00:60:69:51:06:b4
Port 2, WWPN 20:02:00:60:69:51:06:b4
Port 3, WWPN 20:03:00:60:69:51:06:b4
...
Port 10, WWPN 20:0a:00:60:69:51:06:b4
Port 11, WWPN 20:0b:00:60:69:51:06:b4
Port 12, WWPN 20:0c:00:60:69:51:06:b4
Port 13, WWPN 20:0d:00:60:69:51:06:b4
Port 14, WWPN 20:0e:00:60:69:51:06:b4
Port 15, WWPN 20:0f:00:60:69:51:06:b4
Each iSCSI entity has an iSCSI node name. It is a logical name based on the several components: the date when the naming authority aquired the doman name, the name of the naming authority, and a unique device name. It is not linked to an IP address. Only targets and initiators are iSCSI entities. Switches, routers, and ports are TCP/IP devices only and do not have iSCSI node names.
An iSCSI node name has the following format:
iqn.yyyy-mm.backward_naming_authority:unique_device_name
yyyy-mm is the month and year in which the naming authority acquired the domain name.
backward_naming_authority is the reverse domain name of the entity responsible for naming this device. An example reverse domain name is com.netapp.
unique_device_name is a free-format unique name for this device assigned by the naming authority.
An iSCSI node name for an initiator that is an application server:
iqn.1987-06.com.initvendor1:sn2345
Each filer has a default node name based on the Network Appliance reverse domain name and the serial number of the filer's NVRAM card.
The default node name for a filer with the serial number 123456789:
iqn.1992-08.com.netapp:sn.123456789
Copyright © 1994-2003, Network Appliance, Inc. All rights
reserved.
Part No. 210-00338
File: c1ov2.htm
Last updated: 01/17/03 17:40:27