Library Table of Contents Index Previous Page (Alt-p) Next Page (Alt-n) Comments

Stage 4: Cabling the switches and nodes at the partner site

Cabling switches at the partner site  Top

You need to cable the switches at the partner site so that they can work with the local switches.

To cable the partner site switches, complete the following steps.

Step

Action

1

Log in to Switch 3 and set it up as described in Stage 2: Configuring the local switches.

2

Repeat Step 1 for Switch 4.

Cabling the remote node and disk shelves at the partner site  Top

To cable the remote node, disk shelves, and switch, complete the following steps, using the applicable diagram for reference.

Step

Action

1

Cable the first Fibre Channel HBA in slot 5 of the remote node to Switch 3 and Switch 4. In the example, the top port of the Fibre Channel HBA is cabled to Port 8 of Switch 3 and the bottom port is cabled to Port 8 of Switch 4.

2

Cable the second Fibre Channel HBA in slot 8 of the remote node to Switch 3 and Switch 4. In the example, the top port of the Fibre Channel HBA is cabled to Port 9 of Switch 3 and the bottom port is cabled to Port 9 of Switch 4.

3

Connect the local node pool 0 disks to the switches by completing the following substeps:

    1. Connect the Input port of the LRC, ESH, or ESH2 module Channel A of disk shelf 3, to port 0 on Switch 4.
    2. Connect the Input port of the LRC, ESH, or ESH2 module Channel B of disk shelf 3, to port 0 on Switch 3.
    3. Connect disk shelf 3 to disk shelf 4 by connecting the Output ports of the LRC, ESH, or ESH2 module of disk shelf 3 to the Input ports of the corresponding LRC, ESH, or ESH2 module of disk shelf 4.
    4. Set the terminate switches on the LRC or ESH modules to Off on all but the last disk shelf in the disk pool, then set the LRC or ESH terminate switches on the last disk shelf to On.

      Note
      The ESH2 modules are self-terminating and therefore do not have a terminate switch.

In the example, the pool 0 disk shelves are connected to switch port 0 on both Switch 3 and 4. However, you can use any open set of switch ports in this quadrant (limited to ports 0-3 in this configuration).

4

Connect the remote node pool 1 disks to the switches by completing the following substeps:

    1. Connect the Input port of the LRC, ESH, or ESH2 module Channel A on disk shelf 1 to port 12 on Switch 4.
    2. Connect the Input port of the LRC, ESH, or ESH2 module Channel B on disk shelf 1 to port 12 on Switch 3.
    3. Connect disk shelf 1 to disk shelf 2 by connecting the Output ports of the LRC, ESH, or ESH2 module of disk shelf 1 to the Input ports of the corresponding LRC, ESH, or ESH2 module of disk shelf 2.
    4. Set the terminate switches on the LRC or ESH modules to Off on all but the last disk shelf in the disk pool, then set the LRC or ESH terminate switches on the last disk shelf to On.

      Note
      The ESH2 modules are self-terminating and therefore do not have a terminate switch.

In the example, the pool 1 disk shelves are connected to switch port 12 on both Switch 3 and 4. However, you can use any open set of switch ports in this quadrant (limited to ports 12-15 in this configuration).

5

Confirm that the disks are visible and have dual paths by entering the following command on the console:

storage show disk -p

Results

The output shows the disks connected to the switches, to what port they are connected, and to what disk and to which LRC, ESH, or ESH2 they belong, as shown in the following example:

Example

ha16*> storage show disk -p
PRIMARY      PORT    SECONDARY       PORT    SHELF  BAY
-------      ----    ---------       ----    -----  --
switch3:0.40   A     switch4:0.40     B        5     0
switch4:0.41   B     switch3:0.41     A        5     1
.
.
.
switch3:12.52  B     switch4:12.52    A        6     4
6

Connect the ISL from Switch 1 to Switch 3 and connect the ISL from Switch 2 to Switch 4.

7

Connect the VI-MC cluster adapter to the switches, with the top port connecting to Switch 3 and the bottom port connecting to Switch 4.

Note
The example shows the VI-MC cluster adapter connecting to switch ports 3 and 13, but you can use any available switch port. For a FAS3000 series MetroCluster, the NVRAM card cannot be used as a cluster adapter and must be moved to slot 1.

8

Boot the node.

9

Check to make sure that the node can see all disk shelves by entering the following command on the console:

aggr status -r
10

Configure the cluster, as described in Chapter 4, Configuring a Cluster.

Library Table of Contents Index Previous Page (Alt-p) Next Page (Alt-n) Comments