Start of call procedure

Provides a starting point for analyzing problems. You should begin all service actions with this procedure.

This is the starting point for diagnosing and repairing servers. From this point, you will be guided to the appropriate information to help you diagnose server problems, determine the appropriate repair action, and then perform the necessary steps to repair the server. A system attention light indicates there is a serviceable event (an SRC in the control panel or in one of the serviceable event views) on the system. This procedure will guide you through finding the serviceable event.

Note: In this topic, control panel and operator panel are synonymous.

Before beginning, perform as many of the following tasks as possible:

Note: Ask the customer for the Using the problem reporting forms (available in the Customer service, support, and troubleshooting topic) that they should have completed when working with the support center.

Verify the symptoms and service call data, including:
- The server machine type, model number, and serial number.
- The customer problem number.
- A reference code (SRC, SRN or progress code). If you do not have a reference code, ask the customer for the date and time of the problem.
- Any available location code information
- Any data stored in:
  - the service action event log in Service Focal Point
  - the Service Action Log (i5/OS® servers)
  - other operating system-specific logs as directed
- Symptoms reported to you by the customer.
- Symptoms reported to you by the service support center.
- Symptoms reported to you by an attached input/output (I/O) device, for example, a disk drive enclosure expansion drawer.
Record information to help you return the server to the same state that the customer typically uses, such as:
- The IPL type that the customer typically uses for the server. (See Function 01: Display selected IPL type, system operating mode, and IPL speed for more information.)
- The IPL mode that is used by the customer on this server.
- The way in which the server is configured or partitioned.
Ensure that the customer has put the server into a state in which you can perform service tasks.

Is the failing component a monitor (display) or keyboard?
No:

Continue with the next step.

Yes:

Is the monitor or keyboard attached to a keyboard, video, mouse (KVM) switch?
No:

Continue with the next step.

Yes:

Go to Troubleshooting the keyboard, video, and mouse (KVM) switch for the 1x8 and 2x8 console manager.
Is the failing unit a 7037-A50 or a 7047-185?
- No: Continue with the next step.
- Yes: Go to Diagnosing a problem on a 7037-A50 or a 7047-185.
Is the failing server an xSeries or a BladeCenter blade server attached to an System i® server through an iSCSI Host Bus Adapter (HBA)?
- No: Continue with the next step.
- Yes: Go to the System i integrated xSeries troubleshooting procedure.
Is there a Hardware Management Console (HMC) attached to the failing unit?
- No: Continue with the next step.
- Yes: Continue with step 7.
Does the customer intend for this system to be an HMC-managed system?
- No: Go to step 9.
- Yes: Continue with the next step.
Attention: See the following considerations to help determine if the server was set up to be an HMC-managed server:
- Ask the customer.
- If the server does not have a control panel, then there should be an HMC connection, and the server is considered an HMC-managed system.
- Look for HMC=x displayed in the control panel. The x represents a 0, 1, or 2 to show the quantity of HMCs that are currently connected to the server.
  Note: If the control panel displays HMC=0, the server was once connected to an HMC and now the HMC is disconnected. Although the HMC is disconnected, the server is configured to be an HMC-managed system. If the customer does not want the system to be HMC-managed, then the customer can go to Restoring your server to factory settings and follow the procedure to restore the server to the factory default settings.
  If the server has been restored to factory settings and the server is still displaying error symptoms, continue with 9.
Inform the customer that an HMC is required to continue servicing the system, and ask the customer to reattach the HMC to the server. When the HMC is reattached to the managed system, continue with the next step.
Is the HMC functional?
- No: Go to HMC problem isolation procedures. Once the HMC is functional, return here and continue with the next step.
- Yes: Continue with the next step.
Perform the following steps from the HMC that is used to manage the server. During these steps, refer to the service data that was gathered earlier:
1. In the Navigation Area, open Service Applications.
2. Select Service Focal Point.
3. Select Repair Serviceable Event.
4. On the Select Failing System window, select the managed system that has the problem, and click OK.
5. Scroll through the log and verify that there is a problem with the status of Open to correspond with the customer's reported problem.
  Note: If you are unable to locate the reported problem, and there is more than one open problem near the time of the reported failure, use the earliest problem in the log.
Do you find the reported problem, or an open problem near the time of the reported problem?
- No: Continue with the next step.
- Yes: Select the serviceable event you want to repair, and select Repair from the Selected menu.
  This launches a series of windows that guides you through the steps to repair the serviceable event. The system guides you through one of the two following methods of repair, depending on the type of FRU you need to exchange:
  - An interactive step-by-step process that provides illustrations and video presentations to help you exchange the FRU.
  - A link to the appropriate information center topic that provides instructions to help you exchange the FRU.
  After you complete the repair procedure, the system automatically closes the serviceable event. This ends the procedure.
  
  Note: If the Repair procedures are not available, continue with the next step.
Is there an eight-digit reference code (except Cxxxxxxx) displayed in function 11 on the control panel, or if applicable on the HMC?
Note: If the control panel is blank and nothing is displayed in the upper left corner (for example 01) suspect a power problem, go to the Power problems.
- No: Continue with the next step.
- Yes: Record all reference code data, including the values for functions 11 through 20 (see Collecting reference codes and system information), and note the failing partition. Then, go to the Reference codes topic to further isolate the problem. This ends the procedure.
Notes: If you are working at an HMC, check the HMC for operator panel values by performing the following steps:
1. In the Navigation Area, expand Server and Partition > Server Management.
2. Choose from the following options:
  - To view managed system reference codes:
    1. In the right pane, right-click the system and select Properties.
    2. Select the Reference Code tab to view the codes.
    3. When finished, click Cancel.
  - To view logical partition reference codes:
    1. In the right pane expand the system that contains the partition.
    2. Open Partitions.
    3. Right-click the logical partition and select Properties.
    4. Select the Reference Code tab to view the codes.
    5. When finished, click Cancel.
Is the system managed by the Integrated Virtualization Manager (IVM)?
Note: For information about the IVM, refer to Managing the Integrated Virtualization Manager
- Yes: Go to step 13.
- No: Continue with the next step.
Is an operating system console session available and can you log into the session?
Notes:
1. For servers with logical partitions, the console session must be available for the failing partition.
2. If you suspect a problem is on a device (a disk drive enclosure expansion drawer for example) that is attached to a server or a partition, the console session must be available for the partition to which the device is attached.
3. See 5250 Console in the Managing your server topic for details on accessing a 5250 console session on the HMC.
- Yes: Go to step 14.
- No: Continue with the next step.
Work with the customer to open an operating system console session. Can you successfully open an operating system console session?
- Yes: Go to step 14.
- No: Go to step 16.
Work with the customer to access the IVM web interface if it is not already running. Refer to Connecting to Integrated Virtualization Manager. Can you successfully open the console session for IVM?
- Yes: Continue with the next step.
- No: Go to step 16.
Use the operating system-specific, the IVM-specific, or the Virtual I/O server (VIOS)-specific service tools and system logs to locate a "serviceable event" in the logs, or if there is no serviceable event, use the customer-reported symptom in step 14.b.
1. Ask the customer for the date and time of the problem.
  Note: If you are unable to locate the reported problem and there is more than one open problem near the time of the reported failure, use the earliest failure.
2. Go to the appropriate procedure depending on the operating system, IVM, or VIOS that is reporting a problem.
  - If your server is managed by IVM, go to Collect serviceable events in IVM, and then return here and continue on step 15.
  - If your server or partition is running AIX, go to AIX fast-path problem isolation.
  - If your server or partition is running Linux®, go to Linux fast-path problem isolation.
  - If your server or partition is running i5/OS, go to Using the Service Action Log.
  - If your server or partition is running a Virtual I/O Server partition, go to Virtual I/O Server troubleshooting.
Did you find a reference code associated with this failure near the time of the reported problem?
- No: Continue with the next step.
- Yes: Collect all reference code data and note the failing partition. Go to the Reference codes topic to further isolate the problem. This ends the procedure.
Use the HMC or a web browser to access the Advanced System Management Interface (ASMI). See Accessing the Advanced System Management Interface for details. Then, perform the following steps:
1. Log into the ASMI.
2. Click the plus sign next to System Service Aids.
3. Select Error/Event Logs to view the service processor error log.
Note: For more information on using the ASMI, see Managing your server using the Advanced System Management Interface.

Were you able to access ASMI and is there a reference code in the service processor error log that requires service?
- No: Continue with the next step.
- Yes: Collect all reference code data and note the failing partition. Go to the Reference codestopic to further isolate the problem. This ends the procedure.
The following examples show the possible formats of reference codes that display during the power-on process:
- C1xx xxxx codes are displayed during the time after power is connected to the server until the service processor initial power-on process completes (indicated by "01" in the upper-left corner of the physical or logical control panel).
- C2xx xxxx and C7xx xxxx codes are displayed after the power-on sequence is initiated.
- CAxx xxxx codes are displayed while an AIX or Linux partition is initializing its resources.
- xxx, 0xxx, 2xxxx, or C6xx xxxx, C9xx xxxx codes are displayed during the time that the operating system boots and configures resources.
- 888-xxx code
Does the power-on process for the server appear to be stopped, and does function 11 display a reference code of a format shown in the list above?
- No: Go to the Symptom index. This ends the procedure.
- Yes: Collect all reference code data and note the failing partition. Go to the List of progress codes topic to further isolate the problem. This ends the procedure.