Replacing the SAN Volume Controller 2145-CF8 service controller

Use the following information to replace the SAN Volume Controller 2145-CF8 service controller.

To replace the service controller, perform the following steps:

  1. When the service controller, as shown by  1  in Figure 1, is close enough to the USB cable ( 3 ), attach the cable to the service controller, if necessary.
    Figure 1. SAN Volume Controller 2145-CF8 service controller
    SAN Volume Controller 2145-CF8 service controller
    •  1  Service controller
    •  2  Service-controller release buttons
    •  3  USB service-controller cable
  2. Align the service controller with the service controller bay on the SAN Volume Controller 2145-CF8 node, and gently begin to push the service controller into the node.
  3. Continue gently pushing the service controller into the frame until the rear connectors are fully seated and the front metal work of the service controller is up to the frame.
    Important: Ensure that the USB cable is pushed back into the empty drive bay, and ensure that the cable is not trapped above or below the bay.
  4. If you removed the node from the rack, replace the node in the rack, as described in Replacing the SAN Volume Controller in a rack.
  5. If you removed any fibre-channel or Ethernet cables, use the labels you placed on each cable to replace all fibre-channel and Ethernet cables in the same ports from which they were removed.
  6. If you removed the power cords, replace the power cords and the cable-retention brackets, as described in Replacing the cable-retention bracket.
  7. If you pulled out the node to the fully-extended position on the rack, lift the locking levers ( 1  in Figure 2) on the slide rails and push the server  2  all the way into the rack until it clicks into place.
    Figure 2. Raising the SAN Volume Controller 2145-CF8 locking levers of the slide rails of the rack
    Raising the SAN Volume Controller 2145-CF8 locking levers of the slide rails of the rack
  8. Turn on the node.
    Note: The worldwide port names (WWPNs) of the fibre-channel ports are derived from the worldwide node name (WWNN) of the service controller. If you do not perform step 9, you might have to rezone the fibre-channel switches if the switch zoning uses WWPN. You must restart the host systems before they are able to access disks through this node.
  9. If you replaced the service controller as part of concurrent maintenance, you must rewrite the WWNN on the new service controller. If you do not, and the fibre-channel switch zoning uses WWPNs, you cannot add the node back into the cluster until the fibre-channel switches in the SAN are rezoned. Also, the host systems cannot access the fibre-channel ports on that node until the host systems are reconfigured.

    The node finally stops with Validate WWNN in the front panel. Press select and use the left and right buttons until either Disk WWNN or Panel WWNN is displayed. If you use Disk WWNN, which is what the node was using before the service controller was changed, you do not have to reconfigure SAN switches and applications.

    Press and release the down button until Use Disk WWNN is displayed. If your selection is correct, press Select to verify that the WWNN that you set matches the WWNN that you recorded when you removed the service controller.

    If the WWNN for the node is not correct, restore the WWNN by performing the following steps:

    1. Determine the WWNN for the node by examining the vital product data (VPD) of the node or the zoning that is defined in the SAN switch. Record the last five numbers of the WWNN for the node that you are working on.
    2. Start the node by performing a node rescue. See Performing the node rescue.
    3. Change the node WWNN by performing the following steps:
      1. On the front-panel display, press and release the down button until the Node panel is displayed. Then press and release the right button until the Node WWNN: panel is displayed.
      2. Press and hold the down button, press and release the select button, and then release the down button. The display switches into edit mode. Edit WWNN is displayed on line 1. Line 2 of the display shows the last five numbers of the WWNN that is currently set. The first number is highlighted.
      3. Change the highlighted number to match the number from the VPD or from customer zoning. Use the up and down buttons to increase or decrease the numbers. The numbers wrap F to 0 or 0 to F. Use the left and right buttons to move between the numbers.
      4. When the displayed value matches the numbers in the VPD or customer zoning, press and release the select button to accept the numbers. The Node WWNN: panel is displayed and the second line shows the last five numbers of the restored WWNN.

Wait one minute. If Cluster: cluster_name is displayed, the node has rejoined the cluster. If Cluster: does not display, see MAP 5000: Start to determine how to solve this problem or contact the IBM® support center.

Library | Support | Terms of use | Feedback
© Copyright IBM Corporation 2003, 2009. All Rights Reserved.