Preparing for the Procedure - Appliance for Hadoop

Appliance for Hadoop 3 and 4 Platforms Hardware Replacement Guide for Customers
Product
Appliance for Hadoop
Published
March 2018
Language
English (United States)
Last Update
2018-05-09
dita:mapPath
ika1471885266504.ditamap
dita:ditavalPath
hadoop3_4.ditaval.ditaval
dita:id
gjo1471650852867
Product Category
Hardware
Wear an anti-static wrist strap attached to an unpainted metal surface on the rack or chassis when handling parts.
Only trained customer personnel should perform this procedure.
  1. If necessary, log on to https://access.teradata.com and initiate an incident.
  2. Obtain a replacement part as described in Obtaining a Replacement Part.
  3. Contact the system administrator if you do not know the host name or IP address of the designated host CMIC (SMWeb 12.06 and later) or SOV-enabled CMIC (SMWeb 12.05 and earlier).
    In SMWeb 12.05 and earlier, you must access the home page of a CMIC that is SOV-enabled. If the CMIC is not SOV-enabled, most web applications and pages are unavailable. If this occurs, you must navigate to the SMWeb home page of an SOV-enabled CMIC. For more information about the SMWeb home page, see SMWeb online help or the Teradata Server Management Web Services User Guide (B035-5350) available at http://www.info.teradata.com.
  4. Remove the new part from its packaging and place it on a static-free surface.
  5. Record the serial number and/or part number of the new part in the site equipment log.
  6. Using a browser, connect to the Server Management Web Services home page by typing the following in the Address field of the browser:
    CMIC Version Home Page Address
    SMWeb 12.06 and later: http://<hostname or IP address of CMIC or Teradata ServiceConnect connection>:9981/CWA/home.asp
    SMWeb 12.05 and earlier: http://<hostname or IP address of SOV-enabled CMIC>:9980/CWA/home.asp
  7. On the Server Management Web Services home page, click Server Management Client.
  8. If prompted, log on to Server Management Client as a Service User. The Server Management Client software opens.
  9. In the Summary Alerts tab, identify the node containing the failed storage drive:
    1. In the Server Management Client window, click the Summary Alerts tab.
    2. In the Summary Alerts tab, select the summary alert folder containing the failed drive.

      In the Synopsis field, the following alert indicates a failed drive:

      A Node Drive Failure has been detected in SMP.x.x.x.x.x.x
      SMP.x.x.x.x.x.x identifies the failed storage drive. The first x is the system ID, the second x is the collective ID, the third x is the cabinet ID, the fourth x is the chassis ID, the fifth x is the backplane ID, and the sixth x is the drive ID.
      Chassis ID numbers are shown in the cabinet build illustrations.
    3. Record the ID of the failed drive.
  10. In the Tree View, verify that the node containing the failed storage drive matches the ID of the failed drive in the summary alert:
    1. If you have a multi-system configuration, in the Server Management Client Tree View, in the System pull-down menu, select All.


    2. Above the System pull-down menu, click Hide normal chassis status. The node with the failed storage drive appears in the Server Management Client Tree View.
    3. In the Tree View, expand the node with the failed drive.
    4. Expand Disk Backplane.


    5. Select the failed drive to highlight it. indicates the failed drive.
    6. In the Server Management Client window, select View > Properties.
    7. In the Properties window, in the Alias row, find the system ID, collective ID, cabinet ID, chassis ID, backplane ID, and drive ID.
    8. Compare the ID of the failed drive in the Tree View to the ID of the failed drive in the summary alert to verify you are working with the same drive.
    9. In the Properties dialog box, click Close.
  11. Contact the system administrator if you do not know the username and password for the root user of the node with the failed drive.
    You will need the username and password to complete this procedure.
  12. Open a maintenance window. This prevents alerts from being forwarded to Teradata Services at sites that are TVI-connected.
    1. In the Server Management Client window, select the node containing the failed drive.
    2. In the Server Management Client window, select Functions > Maintenance Window. A Maintenance Window dialog box appears.


    3. In the Maintenance Duration field, set the Start Time and the End Time. Start Time is the date and time the maintenance window opens, and End Time is the date and time the maintenance window closes. The default is four hours.
    4. In the Name field, enter a name for the maintenance window.
    5. Click Activate.
    6. In the Information dialog box that appears, click OK to acknowledge the maintenance window has been saved. If more time is needed to complete the procedure, extend the duration of the open maintenance window: change the number of hours in the Maintenance Duration field and click Activate.
  13. Select the node to highlight it.
  14. Click Functions > Locator LED. The Locator LED Enabled window appears. In the cabinet, the LED of the affected node illuminates.
  15. At the front of the cabinet, locate the drive with a solid amber or blinking amber LED.
    Data nodes come in 12-drive and 24-drive versions. See Dell R720xd Chassis Specifications.
    If more than one drive in the same node has failed since the initial summary alert, you can now replace all failed drives in that same node.