Table of ContentsLibraryView in Frames

Configuring SnapMirror over Fibre Channel

To use SnapMirror over Fibre Channel, you must install the Fibre Channel NIC adapters, and set up the SnapMirror relationships. You can also set up multiple SnapMirror traffic zones on supported switches.

Before You Begin

To check for requirements and supported switches for a Fibre Channel connection, refer to Requirements for SnapMirror Over Fibre Channel Transport (asynchronous, synchronous, and semi-synchronous modes) in the Product Documentation section of the NOW site.

Considerations

For updated information about the hardware supported with different storage system models, see the System Configuration Guide.

To configure the systems for Fibre Channel use, complete the following steps.

Steps

  1. Install the Fibre Channel NIC adapters in the source and destination systems.

    See the System Configuration Guide for installation instructions.

  2. Connect the systems to Fibre Channel switches. For more information, see the hardware and service guide for your storage system.
  3. Use the sysconfig command to identify the Fibre Channel NIC adapters on each storage system.

    Example

    The following example shows an output of this command where the storage system has a Fibre Channel NIC adapter installed in slot 4.

    filer_1*> sysconfig
    Release 7.3: Wed Mar 31 02:47:49 PST 2008
    System ID: 0033587346 (filer_1); partner ID 0033586737
    (filer_2)
    System Serial Number: 1022105 (filer_1)
    System Rev: B0
    Backplane Part Number: 0
    Backplane Rev: B0
    Backplane Serial Number: 1022105
    slot 0: System Board
    Processors:1
    Memory Size:3072 MB
    CIOB Revision ID:5
    slot 0:FastEnet-10/100 Ethernet Controller
    e0 MAC Address: 00:a0:98:00:f5:39 (auto-100x-fd-up)
    slot 4: FCVI Host Adapter 4a
    slot 4: FCVI Host Adapter 4b
    slot 6: Gigabit Ethernet Controller IV
    e6 MAC Address 00:02:b3:aa:19:d6: (auto-100x-fd-up)
    slot 7: FC Host Adapter 7
    28 Disks 1904.0GB
    slot 8: FC Host Adapter 8
    28 Disks: 1904.0GB
    slot 10: VI Cluster Adapter
    slot 11: NVRAM
    Memory Size: 128 MB

  4. Enter the sysconfig -v command to display port connection details, including host port ID acquired from the login to the switch, as well as the ID of the switch port to which it connects.

    Example

    The following command displays port connections for the adapter in slot 4.

    filer_1*> sysconfig -v 4
    slot 4: FCVI Host Adapter 4a (Dual Channel, QLogic 2312
    (2352) rev. 2, 64
    -bit,F-port, <UP>
    Firmware rev: 3.1.18
    Host Loop Id:0xffHost Port Id: 0xa01200
    Cacheline size:8FC Packet Size: 2048
    SRAM parity:yesExternal GBIC: No
    Link Data Rate:2 Gbit
    Switch Port brcd_sw_1: 2
    slot 4: FCVI Host Adapter 4a (Dual Channel, QLogic 2312
    (2352) rev. 2, 64
    -bit,F-port, <UP>
    Firmware rev: 3.1.18
    Host Loop Id:0xffHost Port Id: 0xa01200
    Cacheline size:8FC Packet Size: 2048
    SRAM parity:yesExternal GBIC: No
    Link Data Rate:2 Gbit
    Switch Port brcd_sw_2: 2

  5. Determine the zoning configuration. See the documentation for your switch configuration information. Ensure that the Fibre Channel switches are in fabric mode and support the Simple Name Service (SNS) protocol with support for symbolic names.

    Steps 6 through 9 show you how to create new zones and a multipath setup between a source storage system and a destination storage system with a Brocade switch. Each storage system has one X1024 Fibre Channel NIC card installed. The port WWNs are listed in the following table.

    Source storage system Destination storage system
    port a: 20:00:00:e0:8b:0a:aa:6d port a: 20:00:00:e0:8b:14:70:af
    port b: 20:01:00:e0:8b:2a:aa:6d port b: 20:01:00:e0:8b:34:70:af

    The primary path is formed by “port a” on the source and destination, and the secondary path is formed by “port b” on the source and destination. SnapMirror traffic needs a dedicated zone, and it is strongly advised that the user have separate zones for the two paths.

    Steps 10 through 13 show you how to create new zones and a multipath setup with a Cisco switch. The zoning ideas are the same when using Cisco or Brocade switches. However, the command semantics are different. Use the config command to enter the switch configuration mode and set up the zoning configuration as shown in Steps 14 through 17.

    Steps 14 through 17 show you how to create new zones and a multipath setup with a McDATA switch. Use the config command to enter the switch configuration mode.

  6. (With a Brocade switch) Create a new zone with the zonecreate command.

    Example

    brcd_sw_1:root> zonecreate "sm_zone_1", "20:00:00:e0:8b:0a:aa:6d; 20:00:00:e0:8b:14:70:af"
    brcd_sw_1:root> zonecreate "sm_zone_2", "20:01:00:e0:8b:2a:aa:6d; 20:01:00:e0:8b:34:70:af"

  7. Create a new zoning configuration with the cfgcreate command.

    Example

    brcd_sw_1:root> cfgcreate "sm_zone_cfg", "sm_zone_1; sm_zone_2"

  8. Enable the zoning configuration with the cfgenable command.

    Example

    brcd_sw_1:root> cfgenable "sm_zone_cfg"zone config "sm_zone_cfg" is in effect
    Updating flash ...

  9. Check the zoning configuration using the cfgshow command.

    Example

    brcd_sw_1:root> cfgshow
    Defined configuration:
    cfg: sm_zone_cfg
    sm_zone_1; sm_zone_2
    zone: sm_zone_1
    20:00:00:e0:8b:0a:aa:6d; 20:00:00:e0:8b:14:70:af
    zone: sm_zone_2
    20:01:00:e0:8b:2a:aa:6d; 20:01:00:e0:8b:34:70:af
    Effective configuration:
    cfg: sm_zone_cfg
    zone: sm_zone_1
    20:00:00:e0:8b:0a:aa:6d
    20:00:00:e0:8b:14:70:af
    zone: sm_zone_2
    20:01:00:e0:8b:2a:aa:6d
    20:01:00:e0:8b:34:70:af

  10. (With a Cisco switch) Define the two zones with the following commands.

    Example

    cisco_sw_1(config)# zone name sm_zone_1
    cisco_sw_1(config-zone)# member pwwn 20:00:00:e0:8b:0a:aa:6d
    cisco_sw_1(config-zone)# member pwwn 20:00:00:e0:8b:14:70:af
    cisco_sw_1(config-zone)# zone name sm_zone_2
    cisco_sw_1(config-zone)# member pwwn 20:01:00:e0:8b:2a:aa:6d
    cisco_sw_1(config-zone)# member pwwn 20:01:00:e0:8b:34:70:af

  11. Define the zoning configuration with the following commands.

    Example

    cisco_sw_1(config)# zoneset name sm_zone_cfg
    cisco_sw_1(config-zoneset)# member sm_zone_1

  12. Activate the zoning configuration with the following commands.

    Example

    cisco_sw_1(config-zoneset)# zoneset activate name sm_zone_cfg
    Zoneset activation initiated. check zone status
    cisco_sw_1(config-zoneset)# member sm_zone_2

  13. Check zoning configuration status with the following commands.

    Example

    cisco_sw_1# show zoneset active
    zoneset name sm_zone_cfg
    zone name sm_zone_1
    pwwn 20:00:00:e0:8b:0a:aa:6d
    pwwn 20:00:00:e0:8b:14:70:af
    zone name sm_zone_2
    pwwn 20:01:00:e0:8b:2a:aa:6d
    pwwn 20:01:00:e0:8b:34:70:af

  14. (With a McDATA switch) Define the two zones with the following commands.

    Example

    mcdata_sw_1> config
    Config> zoning addzone sm_zone_1
    Config> zoning addzone sm_zone_2
    Config> zoning addwwnmem sm_zone_1 20:00:00:e0:8b:0a:aa:6d
    Config> zoning addwwnmem sm_zone_1 20:00:00:e0:8b:14:70:af
    Config> zoning addwwnmem sm_zone_2 20:01:00:e0:8b:2a:aa:6d
    Config> zoning addwwnmem sm_zone_2 20:01:00:e0:8b:34:70:af

  15. Rename the zone set to "sm_zone_cfg".

    Example

    Config> zoning renamezoneset sm_zone_cfg

  16. Activate the zoning configuration.

    Example

    Config> zoning activatezoneset

  17. Check the zoning configuration status.

    Example

    mcdata_sw_1> show zoning
    Default Zone Enabled: False
    Zone Set: sm_zone_cfg
    Zone: sm_zone_2
    Zone Member: 20:01:00:E0:8B:34:70:AF
    Zone Member: 20:01:00:E0:8B:2A:AA:6D
    Zone: sm_zone_1
    Zone Member: 20:00:00:E0:8B:14:70:AF
    Zone Member: 20:00:00:E0:8B:0A:AA:6D

  18. Determine the IP address and net mask for each port. If you have a active/active configuration, you also need to decide the IP address of the partner port. These IP addresses must be within the private network IP address range.

    Note: You should configure different private subnets for each Fibre Channel port on the storage system.

  19. Use the setup command to configure the IP address. This ensures that the changes are committed to non-volatile storage and persist after a storage system reboot.

    Example

    The following example shows an abbreviated setup command output displayed when configuring a Fibre Channel NIC adapter.

    filer_1*> setup
    The setup command will rewrite the /etc/rc, /etc/exports, /etc/hosts, /etc/hosts.equiv, /etc/dgateways, /etc/nsswitch.conf, and /etc/resolv.conf files, saving the original contents of these files in .bak files (e.g. /etc/exports.bak).
    Are you sure you want to continue? [yes] yes
    
    Release 7.3: Wed Mar 31 02:47:49 PST 2006
    System ID: 0033587346 (filer_1); partner ID: 0033586737 (filer_2)
    	System Serial Number: 1022105 (filer_1)
    	System Rev: B0
    	Backplane Part Number: 0
    	Backplane Rev: B0
    	Backplane Serial Number: 1022105
    	slot 0: System Board
    		Processors: 1
    		Memory Size: 3072 MB
    		CIOB Revision ID: 5
    	slot 0: FastEnet-10/100 Ethernet Controller
    		e0 MAC Address: 00:a0:98:00:f5:39 (auto-100tx-fd-up)
    slot 4: FCVI Host Adapter 4b
    	slot 4: FCVI Host Adapter 4b
    	slot 6: Gigabit Ethernet Controller IV
    		e6 MAC Address: 00:02:b3:aa:19:d6 (auto-1000sx-fd-up)
    	slot 7: FC Host Adapter 7
    		28 Disks: 1904.0GB
    		2 shelves with LRC
    	slot 8: FC Host Adapter 8
    		28 Disks: 1904.0GB
    		2 shelves with LRC
    slot 10: VI Cluster Adapter
    	slot 11: NVRAM
    		Memory Size: 128 MB
    Please enter the new hostname [filer_1]:
    Do you want to configure virtual network interfaces? [n]: . . . .
    Please enter the IP address for FC Network Interface ql4a [10.1.1.15]: 10.1.1.15
    Please enter the netmask for FC Network Interface ql4a [255.255.255.0]: 255.255.255.0
    Should interface ql4a take over a partner IP address during failover? [y]: y
    Please enter the IP address or interface name to be taken over by ql4a [10.1.1.16]: 10.1.1.16
    Please enter the IP address for FC Network Interface ql4b [10.1.2.15]: 10.1.2.15
    Please enter the netmask for FC Network Interface ql4b [255.255.255.0]: 255.255.255.0
    Should interface ql4b take over a partner IP address during failover? [y]: y
    Please enter the IP address or interface name to be taken over by ql4b [10.1.2.16]: 10.1.2.16 . . . .

    You need to reboot the storage system for the changes to take effect. Use the reboot command.

  20. Reboot the storage systems and Fibre Channel switches.
  21. Use the ifconfig command to verify the IP address configuration.

    Note: Ensure that IP addresses for the ports on the same fabric have the same net number.
    A port on the FC NIC adapter is named as qlxa or qlxb, where:
    • ql—Indicates the card vendor. (At this point it is ql only, which stands for QLogic)
    • x—The slot number in which the card is installed in the storage system
    • a/b—The FC port number on the card

    Example

    The followin example shows an output for the ifconfig command.

    filer_1*> ifconfig ql4aql4a: flags=840041<UP,RUNNING,LINK_UP> mtu 8160
    	inet 10.1.1.15 netmask 0xffffff00 broadcast 0.0.0.0
    	partner inet 10.1.1.16 (not in use)
    	ether 00:00:00:00:00:00 (VIA Provider)
    filer_1*> ifconfig ql4bql4b: flags=840041<UP,RUNNING,LINK_UP> mtu 8160
    	inet 10.1.2.15 netmask 0xffffff00 broadcast 0.0.0.0
    	partner inet 10.1.2.16 (not in use)
    	ether 00:00:00:00:00:00 (VIA Provider)
    Note: Under normal operation, you should see UP and LINK_UP in the command output. UP means that this interface has been enabled and is working. LINK_UP means that the physical connection of the interface is online.

  22. Use the route command to verify the IP routing table setup. As this is on a private network, it has a special entry in the routing table.

    Example

    filer_1*> route -sRouting tables
    
    Internet:
    Destination		Gateway		Flags		Refs		Use Interface
    default harley-29-19.lab.n UGS 4		319989		e0		10.1.1/24 filer_1-ql4a U 0 0 ql4a
    10.1.2/24 filer_1-ql4b U 0 0 ql4b
    127 localhost UGS 0 0 lo
    localhost localhost UH 1 0 lo

    The preceding entries specify that for IPs within subnet 10.1.1.x, use the ql4a interface; for IPs within subnet 10.1.2.x, use the ql4b interface.

  23. Use the ping command to verify the connectivity between two network interfaces.

    Example

    filer_1*> ping -s 10.1.1.26
    
    64 bytes from 10.1.1.26 (10.1.1.26): icmp_seq=0 ttl=255 time=0.903 ms
    64 bytes from 10.1.1.26 (10.1.1.26): icmp_seq=1 ttl=255 time=0.462 ms
    64 bytes from 10.1.1.26 (10.1.1.26): icmp_seq=2 ttl=255 time=0.439 ms
    64 bytes from 10.1.1.26 (10.1.1.26): icmp_seq=3 ttl=255 time=0.442 ms
    --- 10.1.1.26 ping statistics ---
    4 packets transmitted, 4 packets received, 0% packet loss
    round-trip min/avg/max = 0.439/0.561/0.903 ms
    Note: After the zoning configuration has been changed, you need to verify the connectivity again.

Related information
http://now.netapp.com