Troubleshooting Fixed-Configuration Router Hardware Faults

Symptom

The STAT indicators on the two MPUs are steady off or blinking green.

Common Causes

• The two MPUs are not properly installed.
• Memory is unavailable.
• Read and write operations on the CF cards fail.
• The system software package is not stored on the CF cards.

Troubleshooting Procedure

1. Check that the MPUs are properly installed.

   Connect the console port of the MPU to the COM port of a PC. Power on the device. If full in or POST failed is displayed in the console port information, check whether the MPU is properly installed.

   Press Ctrl+B to enter bootload Menu…0
   reset mode reg value = 0x7
   reset mode = 0x2
   full in . . . . . . . . . . . . . . . . . . . . . . . .fail
    
   dev_name: eTSEC1
   waiting for auto negotiation to complete... 
   Speed 1000, full duplex
   recv seq num:1
   !!!POST failed, reset board... 

   If the MPUs are properly installed, go to step 2.

2. When the message "Press CTRL+T for full memory test" is displayed in the console port information, press Ctrl+T immediately to fully check the memory. If pass is displayed after the check, the memory is normal. In this case, go to step 3. Otherwise, the memory is faulty. In this case, replace the MPUs.
3. Check that the CF card works properly.

   Generally, if the following information is displayed, the CF card has encountered a read/write fault. In this case, replace the MPUs.

   scan usb storage failed, reset usb and try again fail_count = 1
   USB: Register 10011 NbrPorts 1
   USB EHCI 1.00
   scanning bus for devices... 1 USB Device(s) found
   scanning bus for storage devices... 
   0 Storage Device(s) found

   If the CF card has no read/write fault, go to step 4.

4. Check that the system software package exists on the CF card.

   When the message "Press Ctrl+B to enter bootload Menu" is displayed in the console port information, press Ctrl+B immediately. On the displayed menu, enter 6 to select List file in CFcard to check the CF card list file. If there is a .cc file in the file list, the system software package exists on the CF card. Otherwise, it does not. In this case, go to step 5.

   Bootload Menu(Hiboot Version: 04.00)
   1. Boot with default mode
   2. Boot from CFcard
   3. Enter ethernet submenu
   4. Set boot file and path
   5. Modify boot ROM password
   6. List file in CFcard
   7. Modify System and Chassis Parameters
   8. Modify start mode
   9. Clear password for console user
   10. Reboot
   11. Enter TPM Submenu
   Enter your choice(1-11): 6

5. Contact Huawei and provide the following information:
   • Results of the troubleshooting procedure
   • Configuration, log, and alarm files

Symptom

The STAT indicator on the slave MPU is steady off or blinking green, and the display device command output shows that the status of the slave MPU is Abnormal and Unregistered.

Common Causes

• The slave MPU is starting.
• The system software packages loaded to the master and slave MPUs differ.
• The slave MPU is not properly installed.
• The memory on the slave MPU fails.
• A system software fault occurs.

Troubleshooting Procedure

Symptom

The STAT indicator on the master MPU is steady red, the STAT indicator on a PIC is steady yellow or off, and the display device command output shows that the status of the PIC is Abnormal and Unregistered.

Common Causes

• The PIC is in the startup period.
• The PIC is not properly installed.
• The PIC is faulty.
• The PIC model is not supported on the device.

Troubleshooting Procedure

Symptom

Services are affected, and a historical board restart alarm is reported to the NMS.

Common Causes

• The board is restarted using a command or is powered off.
• The software or hardware of the board is faulty.

Troubleshooting Procedure

1. Run the display board-reset slot-id command in the diagnostic view to check the restart cause.

   <HUAWEI> system-view 
    [~HUAWEI] diagnose 
    [~HUAWEI-diagnose] display board-reset 9 
    Board 9 reset information:
    -- 1. DATE:2018-01-21 TIME:03:20:52+04:00 BARCODE:030QAF10D7000030 RESET Num:0
    -- Reason:Board register, BarCode is 030QAF10D7000030. 
    -- BootMode:NORMAL
    -- BootCode:0x060100ff 

2. Handle the problem using the cause-specific operation recommended in table below.
   Table 1-1 Board restart causes and troubleshooting instructions

   Cause	Description	Recommended Operation
   User operations
   Power off board from command.	A user has reset the board using a command or through the NMS.	Check whether the board is restarted using a command or from being powered off.
   Reset board from PIC command.
   Reset the chassis from command.
   Canbus request to power off the board.
   System loading
   EPLD is upgrade,and reset board.	The board restarts after the EPLD is loaded.	No operation is required.
   Board update by JTAG, and reset board.	The board restarts after it is upgraded using the JTAG channel.
   Board update mbus, and reset board.	The board restarts after the MBus is upgraded.
   Software exceptions
   Board task exception occurs and reset lpu.	The system detects a software exception.	Collect user logs, diagnostic logs, and the display service-diagnostic-information command output, and contact Huawei.
   Board task deadloop occurs and reset lpu.	The system detects a deadloop on the board.	Collect user logs, diagnostic logs, and the display service-diagnostic-information command output, and contact Huawei.
   Component report failure.	The component fails to be reported.	Run the display reportfailure show_num [ begin_num ] [ verbose ] [ all | slave | slot slot-id ] command to check the cause. Collect user logs, diagnostic logs, and the display service-diagnostic-information command output, and contact Huawei.
   Multiple-bit ECC check error, and reset board.	A soft error occurs on the chip.	Collect user logs, diagnostic logs, and the display service-diagnostic-information command output, and contact Huawei.
   Device management
   The heartbeat lost and reset board.	The MPU does not receive heartbeat messages from the PIC.	Collect user logs, diagnostic logs, and the display service-diagnostic-information command output, and contact Huawei.
   Semls register failed, and reset board.	A board remains unregistered for a long time.	Collect the console port information recorded during board startup and contact Huawei.
   Hardware components
   Slave mpu is not compatible with the master, and power off board.	Hardware is incompatible.	Replace the slave MPU so that two MPUs are compatible.
   This version does not support the type of board, and power off the board.		Remove the PIC and replace it with a PIC supported by the device and slot. For details about the PIC models supported by the slots, see "Boards" in the Hardware Description of your product.
   Board is incompatible with chassis, and power off board.
   Board is incompatible with slot, and power off board.

Symptom

The optical module information cannot be queried using the display interface command, or the electronic label information of an optical module cannot be queried using the display elabel optical-module interface interface-number command.

Common Causes

• The optical module is not Huawei certified.
• The optical module is not properly installed.
• The optical module or device encounters exceptions.

Troubleshooting Procedure

Symptom

The actual transmit power of an optical module, as measured by an optical power meter, is lower than the nominal transmit power of the module. The display interface command output shows that the transmit power of the optical module is lower than the alarm threshold.

Common Causes

• The optical interface is dirty.
• The optical module was faulty.

Troubleshooting Procedure

1. Check whether the optical interface is dirty. If it is, use a cotton swab to gently clean it. Then, re-test the transmit power. Use dust-proof caps to protect unused optical modules.
2. If the transmit power of the optical module is still abnormal, remove the optical module and install it on another optical interface. If the fault persists, the optical module is faulty. Replace the optical module and send back the faulty one for repair or contact Huawei.

Symptom

The peer NMS reports a CRC bit error alarm, and Output in the display interface interface-type interface-number extensive command output shows that excessive error packets are dropped on the local interface.

Common Causes

• The optical fiber encounters high connector loss or large bend radius.
• The optical module was faulty.

Troubleshooting Procedure

1. Use an optical power meter to measure the receive power of the optical module and check whether it is the consistent with the receive power described in the product manual.
   NOTE:

   To measure the receive power of a fiber, place the optical power meter close to its receive end.

2. If the receive power is below the threshold, check whether the fiber link is normal. Replace the optical fiber and check whether the peer end still reports the bit error alarm.
3. If the problem persists, remove the optical module and install it on another optical interface. If the problem persists, the optical module is faulty. Return the optical module for repair or contact Huawei.

Symptom

A device is connected to a remote device through optical interfaces and fibers. The two interfaces are down and the devices cannot communicate with each other.

Common Causes

• The optical module used on the device is not Huawei certified.
• The optical module and fiber types are incompatible.
• One or both of the optical interfaces were shut down.
• The configurations on the local and remote interfaces are inconsistent.
• The transmit power of the optical module is too low or high.
• The receive power of the optical module is too low or high.
• The optical module fails.

Troubleshooting Procedure

1. Check that the optical module is Huawei certified.

   Using Huawei-certified optical modules on the device is recommended because other modules cannot provide required reliability and may prevent the interface from going up.

   NOTE:

   You can check whether a module is Huawei certified by checking its label. If the label has a Huawei logo, the optical module is Huawei certified. Otherwise, send the optical module model to Huawei to check whether it is certified.

2. Check that the optical module matches the optical fiber.

   If they do not match, replace one of them.

   • A single-mode optical module (typically with a center wavelength of 1310 nm or 1550 nm) must be used with single-mode optical fibers (typically yellow).
   • A multimode optical module (typically with a center wavelength of 850 nm) must be used with multimode optical fibers (typically orange).
3. Check that the interface configurations of the local and peer devices are the same.

   Run the display this interface command on both interfaces to check the bandwidth and auto-negotiation mode of both ends. If the bandwidth or auto-negotiation modes differ, modify the configuration to be consistent.

4. Check that the optical module works properly.

   Run the display optical-module { extend | base } information interface { interface-type | interface-number } or display optical-module { extend | base } information slot slot-id pic pic-id command to check the working status of the optical module. Run the display alarm hardware command to check the alarm information of the optical module..

   <HUAWEI> system-view
   [HUAWEI] diagnose
   [HUAWEI-diagnose] display optical-module extend information interface GigabitEthernet 0/7/0
   ============================================================================== 
   Transceiver Digital Diagnostic Monitoring (DDM), Externally Calibrated
   =============|================================================================= 
   Card7-Port0 + Value HighAlarm HighWarn LowWarn LowAlarm Status
    -------------|-----------------------------------------------------------------
    Temperature(C) 48.500 80.000 75.000 -5.000 -10.000 Normal
    Supply Voltage(V) 3.293 3.700 3.630 2.970 2.850 Normal
    Tx Bias(mA) 21.600 58.091 48.091 4.762 4.762 Normal
    Tx Power(avg dBm) -6.106 -3.103 -4.102 -8.102 -9.104 Normal
    Rx Power(avg dBm) -4.025 -3.000 -3.000 -19.030 -19.030 Normal
   =============|=================================================================
   
    
    
   <HUAWEI> display alarm hardware
   Index  Level    Date       Time      Info
   
   1      critical 2018-10-27 14:58:18  GigabitEthernet0/4/4 is failed, the optica
                                        l module on card was removed
   ------------------------------------------------------------------- 

   In the optical module alarm information, check the alarm causes.

   • Low receive power

     The strength of received signals is too low. A possible reason is that the remote interface is down or packets sent from the remote interface are dropped during transmission. In this case, check whether the distance between the two devices exceeds the maximum transmission distance of the optical module. If it does not, check whether the optical module on the remote interface or the optical fiber is damaged. If one is damaged, replace it.

   • High receive power

     The strength of received signals is too high. A possible reason is that the local and remote ends are nearby but a long-distance optical module is used on the remote end. In this case, install an optical attenuator on the remote module to reduce the transmit power.

   • Low transmit power

     The strength of signals sent from the local optical module is too low, or the optical module is faulty. The receive power on the remote end may be low, preventing the remote interface from going up, or causing the interface to drop packets. In this case, contact Huawei.

   • High transmit power

     The strength of signals sent from the local optical module is too high. This may cause a high receive power on the remote optical module. High receive power for a long time may burn the remote optical module. Another possible cause is that the local optical module has failed. In this case, replace the optical module.

     To ensure normal communication between two optical interfaces, after the two interfaces are connected using optical modules and optical fibers, check for transmit and receive power alarms. Ensure that the transmit and receive power values of the two optical modules are within the normal ranges. Otherwise, traffic forwarding on the optical interfaces may be abnormal and the optical modules may become damaged.

5. If neither end has any alarms reported, use a single fiber to loopback the interface.
   NOTE:

   Optical fibers no longer than 10 km can be directly used for loopback. Longer optical fibers require optical attenuators for loopback.

   • If the looped-back interface does not go up, use another optical module and retry the loopback. If the problem persists, the device is faulty. In this case, contact Huawei.
   • If the looped-back interface goes up, check whether the peer device and link are faulty.
6. Replace the optical fiber or optical module and check whether the interface goes up. If the interface goes up, the original optical fiber or optical module is faulty. In this case, replace the faulty fiber or module. Otherwise, contact Huawei.