Clear the Fault and Verify the Functionality of the Replacement DIMM

This procedure describes how to clear a memory fault and how to verify the functionality of the replacement DIMM.

Before You Begin

Ensure that the following conditions are met:

• The server module is in Standby mode (installed in a powered modular system, but the server module’s host is not started).

  See Prepare the Server Module for Removal.

• You have connectivity to the SP.

  See Access the SP (Oracle ILOM)

1. Access the Oracle ILOM -> prompt.

   See Access the SP (Oracle ILOM).

2. Determine how to clear the fault.

   The method you use to clear a fault depends on how the fault is identified by the show faulty command.

   Examples:

   • If the fault is a host-detected fault (displays a UUID), continue to Step 3. For example:

     -> show faulty
     Target              | Property               | Value
     --------------------+------------------------+--------------------------------
     /SP/faultmgmt/0     | fru                    | /SYS/MB/CMP0/BOB0/CH0/D0
     /SP/faultmgmt/0     | timestamp              | Dec 14 22:43:59
     /SP/faultmgmt/0/    | sunw-msg-id            | SUN4V-8000-DX
     faults/0            |                        |
     /SP/faultmgmt/0/    | uuid                   | 3aa7c854-9667-e176-efe5-e487e520
     faults/0            |                        | 7a8a
     /SP/faultmgmt/0/    | timestamp              | Apr 24 22:43:59
     faults/0            |                        | 

   • If the fault was detected by POST and resulted in the DIMM being disabled, you will see something similar to the following output:

     -> show faulty
     Target              | Property               | Value
     --------------------+------------------------+--------------------------------
     /SP/faultmgmt/0     | fru                    | /SYS/MB/CMP0/BOB1/CH0/D0
     /SP/faultmgmt/0     | timestamp              | Apr 24 16:40:56
     /SP/faultmgmt/0/    | timestamp              | Apr 24 16:40:56 faults/0
     /SP/faultmgmt/0/    | sp_detected_fault      | /SYS/MB/CMP0/BOB1/CH0/D0
     faults/0            |                        | Forced fail(POST)

   In most cases, the replacement of the faulty DIMM is detected when the SP is power cycled. In this case, the fault is automatically cleared from the system. If the fault is still displayed by the show faulty command, then use the set command to enable the DIMM and clear the fault.

   Example:

   -> set /SYS/MB/CMP0/BOB0/CH0/D0 component_state=Enabled

3. Perform the following steps to verify the repair:
   1. Set the virtual keyswitch to diag so that POST will run in Service mode.

      -> set /SYS/keyswitch_state=Diag
      Set ‘keyswitch_state’ to ‘Diag’

   2. Power cycle the system.

      -> stop /SYS
      Are you sure you want to stop /SYS (y/n)? y
      Stopping /SYS
      -> start /SYS
      Are you sure you want to start /SYS (y/n)? y
      Starting /SYS 

      ---

      Note - The server module takes about one minute to power off. Use the show /HOST command to determine when the host has been powered off. The console will display status=Powered Off.

      ---

   3. Switch to the system console to view POST output.

      -> start /HOST/console

      Watch the POST output for possible fault messages. The following output is a sign that POST did not detect any faults:

      .
      .
      .
      0:0:0>INFO:
      0:0:0>    POST Passed all devices.
      0:0:0>POST:    Return to VBSC.
      0:0:0>Master set ACK for vbsc runpost command and spin...

      ---

      Note - Depending on the configuration of Oracle ILOM variables that affect POST and whether POST detected faults or not, the system might boot, or the system might remain at the ok prompt. If the system is at the ok prompt, type boot.

      ---

   4. Return the virtual keyswitch to Normal mode.

      -> set /SYS keyswitch_state=Normal
      Set ‘ketswitch_state’ to ‘Normal’

   5. Switch to the system console and type the Solaris OS fmadm faulty command.

      # fmadm faulty

      No memory faults should be displayed.

      If faults are reported, refer to the Diagnostics Process for an approach to diagnose the fault.

4. Switch to the Oracle ILOM prompt (->).
5. Type the show faulty command.
   • If the fault was detected by the host and the fault information persists, the output will be similar to the following example:

     -> show faulty
     Target              | Property               | Value  
     --------------------+------------------------+-------------------------------
     /SP/faultmgmt/0     | fru                    | /SYS/MB/CMP0/BOB0/CH0/D0
     /SP/faultmgmt/0     | timestamp              | Dec 14 22:43:59
     /SP/faultmgmt/0/    | sunw-msg-id            | SUN4V-8000-DX
     faults/0            |                        |
     /SP/faultmgmt/0/    | uuid                   | 3aa7c854-9667-e176-efe5-e487e520
     faults/0            |                        | 7a8a
     /SP/faultmgmt/0/    | timestamp              | Dec 14 22:43:59                 
     faults/0            |                        |

   • If the show faulty command does not report a fault with a UUID, the fault is cleared. You do not need to proceed with the following steps.
6. (Only if previous steps did not clear the fault) Type the set command.

   -> set /SYS/MB/CMP0/BOB0/CH0/D0 clear_fault_action=true
   Are you sure you want to clear /SYS/MB/CMP0/BOB0/CH0/D0 (y/n)? y
   Set ’clear_fault_action’ to ’true’

7. (Only if previous steps did not clear the fault) Switch to the system console and type the fmadm repair command with the UUID.

   Use the same UUID that was displayed from the output of the Oracle ILOM show faulty command.

   # fmadm repair 3aa7c854-9667-e176-efe5-e487e5207a8a

Related Information

• Install a Replacement DIMM

• Verify DIMM Functionality