subscribe iconSubscribe to this information

Planning checklist for concurrent node repair (9117-MMA and 9406-MMA)

Use this checklist as a guide to complete the planning requirements for a concurrent node repair for the 9117-MMA and 9406-MMA. The checklist contains a list of planning tasks that must be completed before you concurrently repair a node, and it also lists who is responsible for each task.

For guidelines that apply to all concurrent maintenance operations, see Guidelines for CEC concurrent maintenance operations. These general guidelines must be adhered to as well as the items in the following checklist. For additional planning considerations, see the IBM® Power 595 and Power 570 Servers CEC Concurrent Maintenance whitepaper (http://www.ibm.com/common/ssi/fcgi-bin/ssialias?infotype=SA&subtype=WH&appname=STGE_PO_PO_USEN&htmlfid=POW03023USEN&attachment=POW03023USEN.PDF).

The term service representative refers to your IBM service representative.

If the responsibility states customer, the task must be completed by the customer before the IBM service representative arrives.

Table 1. Concurrent node repair checklist for the 9117-MMA and 9406-MMA
Concurrent node repair tasks Responsibility

__

Verify that the system firmware level EM340_061 (Service Pack 3.4.1), or later, is activated on the system. Firmware level EM340_xxx is required for a concurrent repair if resources in the node are active and being used by the system. For information about determining firmware levels using the Hardware Management Console (HMC), see Using the HMC to view the existing firmware levels.

Customer

Service representative must verify

__

Verify that the HMC level V7R3.4.0 (Service Pack 1), or later, is installed on the HMC that manages the system. If there are two HMCs attached to the system, both HMCs must be at the same level. If not, the HMC that is not at the required level must be correctly disconnected from the managed system and powered off. For information about verifying the HMC code level and release, see Determining your HMC code level and release . For information about disconnecting an HMC from a managed system, see Removing HMC connection data .

Customer

Service representative must verify

__

If system firmware Ex340_061 (service pack 3.4.1) or Ex340_075 (service pack 3.4.2) is installed on the system, complete the following steps on the HMC.
  1. Go to Systems Management > Servers > target server.
  2. From the list of partitions that are running on the target server, select the first partition name in the list.
  3. In the Partition Properties panel, select Hardware > Processors.
    • If the partition uses shared processors, no action is required.
    • If the partition has dedicated processors, select Allow when partition is inactive under Processor Sharing before you perform a concurrent maintenance operation.
  4. Repeat this procedure for all partitions that are defined on the target system before you begin the concurrent maintenance operation.

Customer

Service representative must verify

__

Verify that the logical memory block size is not set to 16 MB if the server firmware is EM340_061.

If the firmware level is EM340_061 and the logical memory block size is 16 MB, the procedure must be performed with the power off.

Customer

Service representative must verify

__

Verify that the enclosure serial number in the vital product data (VPD) matches the serial number on the enclosure label on the front of each node. If one or more serial numbers does not match, contact IBM to correct it during a maintenance window. Concurrent repair requires a valid enclosure serial number for each drawer in the system.

To verify the enclosure serial number in the system VPD, complete the following steps.
  1. Logon to the HMC, then
  2. Select System Management > Servers > target server
  3. Under Tasks: system name in the lower part of the HMC panel, click Properties.
  4. Click I/O.
  5. Look at the left side in the Slot column. Verify that none of the location codes in the slot column have the format Uxxx.yyy.DQ1234# and that the serial numbers shown match the enclosure serial numbers on the labels. Scroll to the bottom to ensure that there are no invalid enclosure serial numbers and that all serial numbers match.
  6. If any of the location codes have the format Uxxx.yyy.DQ1234#, or do not match, contact IBM to schedule a repair.

Customer

Service representative must verify

__

Verify the network connections between the system's service processors and the HMC before the service representative arrives by using the service processor status function on the HMC. For more information about testing the network connections, see Testing the connection between the HMC and the managed system .

Customer

Service representative must verify

__

Verify that the following program temporary fixes (PTFs) are activated, if IBM i partitions are running on the system.
  • For IBM i 5.4, the MF45678 PTF is required.
  • For IBM i 6.1, the MF45581 PTF is required.
If the PTFs are not activated, the IBM i partitions must be powered off before the concurrent maintenance operation can begin.

Customer

Service representative must verify

__

Ensure that the system is not in power-saver mode. Use the HMC to disable power-saver mode, if necessary. For more information, see Power management .

Customer

Service representative must verify

__

Run the Display Service Effect utility on the HMC to determine if the system is prepared for a concurrent node repair. For more information about the utility, see Displaying the effect of system node evacuation. When using this utility on a node that is deactivated (cold repair), the utility will report no errors. Run the Display Service Effect utility on the HMC to determine if the system is prepared for a concurrent hot node repair.

Customer

Service representative must verify

__

Enable service processor redundancy, except on a 9117-MMA or 9406-MMA with a single node.

Customer

Service representative must verify

__

The system must be quiesced before the concurrent node repair begins. This means that all critical applications must be halted or moved to another system before the operation begins. Noncritical applications can be left running. The partitions may be left running at the operating system command prompt.

Customer

__

Ensure that the replacement parts are the same type as the parts that are being removed.

Service representative

__

Replace only the existing parts. New hardware must not be added during a concurrent node repair.

Service representative

__

Ensure that existing node hardware is not moved during the concurrent node repair.

Service representative


Send feedback | Rate this page

Last updated: Tue, August 16, 2011