Shogun Diagnostic Codes Reference Guide (Rel 1.8.8)

This document lists the diagnostic codes that your server can produce. Some of these codes indicate errors or malfunctions; others are simply way-markers that indicate normal progress. The codes can appear in the server’s front panel LCD display or on the screen in the System Management Application software.

Note
Some codes described in this guide will seem obscure to less experienced users, particularly those codes which appear on the Front Panel LCD. If you cannot understand the meaning of a code, ask an authorised engineer for help.

Codes that appear on the Front Panel LCD

When you press the Power On button from Standby mode, the server initiates several phases in the boot process. During these phases the system performs the following operations:

Checks the Diagnostic Processor.

Checks the System Management Controller (SMC).

Checks the BIOS, which executes the Power On Self Test (POST).

Loads the Operating System.

Attaches the Device Driver and associated clients.

There is a set of diagnostic codes, expressed as hexadecimal numbers, associated with each phase. If your server fails to complete any of these phases, the system will halt and one of the codes will flash on and off on the front panel LCD display. The code represents the point at which the power-on sequence came to a stop and may indicate what element of the system could be at fault.

The appearance of some codes may require an engineer to deal with the problem. In this case, be sure to do the following:

1. Note down the code to help the engineer diagnose the fault.

2. Clear the LCD (and silence the alarm if it is sounding) by using the key to turn the front drive bay door lock to the Unlock position (door closed). If it is already in the Unlock position, turn it to Lock and then to Unlock again.

Power Failure Codes

The LCD will display these codes if the mains electricity supply fails. These codes, shown in brackets below, represent three power failure modes. In each mode, the power supply battery pack powers the system. Each code uses a countdown timer, indicated by nnn, to indicate the number of seconds before its mode expires and the next mode begins. If the countdown begins at 1000 or greater, the LCD will display 999 and hold until the countdown actually reaches 999.

The LCD will display the progress of the countdown from then on.

Brownout (A.nnn) This mode indicates a temporary reduction or even absence of the AC electric power. If it turns out that full power is not restored after approximately five seconds, Brownout will change to Battery Mode. If the battery charge is already low and power consumption is high, the system may skip Battery Mode altogether and jump to Power Fail mode. On the other hand, if power is fully restored before the countdown reaches zero, system function will return to normal and the batteries will recharge.

Battery Mode (b.nnn) This mode results because power has not been restored in Brownout Mode. The system warns users to save their work, terminate their applications and log off the network. System shutdown is initiated in this mode. If power is fully restored before the countdown reaches zero the system will return to normal function and recharge the batteries; otherwise Power Fail mode begins.

Power Fail (F.nnn) In this mode, power has not yet been restored; system shutdown should be in progress to prevent deep battery discharge. Shutdown continues even if power should be restored at this late stage.

 Under normal conditions, the server deals with the power failure itself by means of the Event Manager and, when necessary, initiates system shutdown. You will probably only see the Power Fail code on the LCD if the brownout lasts that long. However, if a brownout occurs and persists when the Event Manager is not loaded, then the LCD will display each of the codes described above. Under these conditions the system is not able to initiate system shutdown. You, or some other person who is local to the server, will need to do it yourself while the system is in Battery Mode. The precise steps you need to take to shut down the system, after instructing network users to close their applications and log off, will depend on the operating system you are using.

When the operating system is at the point where it is safe to switch the machine to Standby, press and hold the Standby (see note below) button on the front panel until the system switches off (see Chapter 2, Operating Your System, in the Owner’s Handbook for more information about switching off). 

System Management Controller (SMC) Error Codes

This section contains SMC error codes. These codes fall into two groups, the range 0F01-0F0F and the range 0F10-0F4F.

0F01-0F0F

These are initialisation codes for the SMC which will only appear immediately after power-up. If the system stops and the LCD flashes one of these codes, it means that there is a serious hardware problem with the SMC board and it probably needs replacing. The following table defines these codes:

0F10-0F4F

These codes indicate various other errors and could appear on the LCD at any point. The following list defines the codes and indicates what action you should take if these errors occur. If an error persists and efforts to solve the problem prove ineffective, you should contact a service engineer.

0F10 Power On Error - This code appears when the firmware does not detect a power-on signal from the Power button within one second. Possible faulty items are the motherboard, the SMC board, the SMIC or the cabling between them. Try pressing the Power button again.

0F11 Power Off Error - Appears when the system will not shut down after pressing the Standby button on the front panel. Possible faulty items are the main circuit board, the SMC board or the cable between them. If it is absolutely necessary to shut down the server, you can switch the circuit breaker switch on the rear panel to the Off position and unplug the mains lead. Only do this, however, as a last resort.

0F12 SMC Watchdog Timer over-flow error - Indicates a faulty SMC board or the SMC firmware. If this code appears, first try resetting the SMC by doing the following:

1. Unlock the removable drive bay door to deactivate the security alarm (see chapter 2 in the Owner’s Handbook).

2. Press the Control, Standby and Reset special buttons simultaneously.

This action displays "8888" on the LCD.

3. Press Standby and Reset simultaneously. This clears the LCD and resets the SMC. If the error code persists, contact a service engineer.

0F13 Too many time-out errors - Indicates faulty SMC board. Try resetting the SMC using the same procedure as for 0F12.

0F15 Invalid interrupt - Indicates a fault in either the SMC board or firmware.

0F20 Diagnostic Processor communication error - Indicates an error in the diagnostic processor on the System Management Interface Card (SMIC), the power distribution board or the cable between them.

0F30 SMC port retry-error occurred - Indicates an error in the SMC board, the SMIC board, the motherboard power distribution board, the cable between them, firmware or the setting of the SMC (or FPSC) variables. First check the settings of the FPSCResponseTimeOut and FPSCReceiptTimeOut variables in the System Management Application (SMA).

0F31 SMC port Response time-out occurred - Same as for 0F30.

0F32 SMC port Transmit time-out occurred - Indicates faulty SMC board or firmware.

0F40 Modem port retry-over occurred - Indicates a faulty modem, modem cable, telephone line, SMC board, settings of SMC (or FPSC) variables in the SMA. You should check the following items first:

1. The modem is connected correctly.

2. The modem is switched on.

3. The modem is connected to the telephone line correctly.

4. The modem works correctly. To double-check this, connect a different modem that you know is in working order.

5. The serial port settings for the modem (e.g. baud rate) are correct.

6. Settings of the MODEMResponseTimeOut and MODEMReceiptTimeOut variables, in the SMA, are correct.

0F41 MODEM port Response time-out - Same as 0F40.

0F42 MODEM port Transmit time-out - Same as 0F40.

0F4D MODEM AT command result code time-out occurred - Indicates faulty modem, modem cable, SMC board or SMC (FPSC) variable settings. You should check the following items first:

1. The modem is connected correctly.

2. The modem is switched on.

3. The modem connection to the telephone is in order.

4. The modem functions properly. To double-check this, connect a different modem that you know is in working order.

5. The serial port settings for the modem (e.g. baud rate) are correct.

6. The settings of the ModemATOriginal, ModemATCommands and MODEMPortBaudRate variables, in the SMA, are correct.

0F4E MODEM AT command failed - Same as 0F4D.

0F4F MODEM AT command unexpected result code is received - Same as 0F4D.

Note	Control Standby & Reset
	Special Button Functions
	Pressing Standby, Control and Reset simultaneously while the front drive bay door is unlocked switches the system into a mode in which these three buttons have special functions. The LCD displays 8888 to indicate this mode.
	STANDBY or RESET - Pressing one of these buttons initiates a memory dump to the central processing unit by activating and then deactivating the Non-Maskable Interrupt (NMI) signal via the diagnostic processor. You can then use the appropriate function of the network operating system to examine the contents of the dump.
	CONTROL - Pressing this button initialises the modem, which is connected to the SMC serial port on the backplane of the Shogun. If the modem initialisation is successful, the LCD displays the code 0000. If the initialisation is unsuccessful, the LCD displays 0F4D or 0F4E.
	STANDBY + RESET - Pressing these buttons simultaneously clears the LCD and then releasing them executes an independent Front Panel System Controller (FPSC) reset. This would only be necessary if a major problem or error had occurred in the system.
	If you do not press any buttons for ten seconds, the system returns to normal mode.

Progress Control (NextBootStage) Codes

A single SMC (FPSC) variable, NextBootStage, records the progress of all phases of the boot process. These phases are:

Power-On Self Test (POST), executed by the BIOS.

SMC Device Driver Attachment (i.e. initialisation).

SMC Client Loading.

The SMC Last Client sends a message to the SMC that it has loaded.

The NextBootStage variable can take values in the range 1000-FFFF. The most significant digits of this variable (i.e. the first three) are progress codes and are allocated as follows:

100-1FF Diagnostic Processor and SMC

200-2FF Motherboard BIOS

300-3FF System Management Interface Card (SMIC) BIOS

400-7FF Unallocated

800-8FF Device Driver

900-EFF Unallocated

F00-FFF Last Client

The least significant digit (i.e. the fourth) indicates whether there is an error condition, according to the following scheme:

0 No error

F Used by SMC to indicate an error

The following codes have special meaning to the SMC:

0001 is issued, accompanied by an alarm, under the following circumstances:

- The nickel-cadmium battery on the SMC board has fully discharged. This will happen if the server is disconnected from the mains electricity for at least one month.

- The SMC firmware which controls the Front Panel is updated.

- A fatal error occurs on the SMC and the SMC executes a self-reset. This means that any changes you have made to the configuration settings within the SMA have been lost. Use the SMA to restore these settings (see the SMA User’s Guide and the SMA’s own on-line help system).

2000 must be issued by the BIOS when it starts execution before configuration testing has taken place.

7FF0 is the POST completion code. It is sent by the BIOS when the POST completes successfully. When the code is received, the POST watchdogs are disabled.

8FE0 is the code sent by the SMC Device Driver to indicate that it will be absent.

8FF0 is the code sent by the SMC Device Driver to indicate that it has initialised successfully.

FFE0 is the code sent by the SMC Last Client to indicate that all clients will be unloaded. The Trap generator is disabled.

FFF0 is the code sent by the SMC Last Client to indicate that all clients are now loaded. The Trap generator is enabled.

SMC-specific codes and NextBootStage codes

POST beep codes

The BIOS signals a failure with a beep code before video adapter initialisation. The code consists of up to 10 beeps of equal length. The following table explains the meaning of each code:

POST codes and Countdown codes

After successful video adapter initialisation, the BIOS indicates the current testing phase during POST by outputting a 2-digit hex code to I/O location 80h. The current countdown code will also be displayed on the LCD panel once it is initialised.

The following table contains the port-80 codes and the POST countdown codes (expressed as hexadecimal numbers) which the system displays during the recovery boot process. During this process the floppy disk in drive A: is booted and a BIOS image is automatically installed.

The following table contains the port-80 and countdown codes displayed during the normal BIOS POST process.

Codes Generated by the SMIC BIOS

The following table contains codes that the System Management Interface Card (SMIC) BIOS generates during POST. The codes, which can appear on the LCD or within the System Management Application, define the start and end points of each action. The appearance of one of these codes does not indicate an error unless it flashes.

Codes that appear in the SMA

POST error codes and messages

The BIOS indicates errors by writing an error code to the PS/2 standard logging area in the Extended BIOS Data Area, and by displaying a message on the screen, which is preceded by the POST Error Code. The error code will also be logged to the Critical Event Logging area. The following codes will appear inside the System Management Application (SMA) and not on the front panel LCD.