Type

Fault

  fault.chassis.i2c.device

Severity

Critical

Description

I2C access failure to an I2C device.

Automated Response

The resources associated with the I2C device may deconfigured. The speed of all fans on the platform may be raised. Please consult the detail section of the knowledge article for additional information.

Impact

It is possible that the platform will not be able to be be restarted. Please consult the detail section of the knowledge article for additional information.

Suggested Action for System Administrator

The platform administrator should schedule a service action to replace the affected Field Replaceable Unit (FRU), the identity of which can be determined using fmdump -v -u EVENT_ID. Please consult the detail section of the knowledge article for additional information.

M4000/M5000 see Doc 1499349.1  FANBP/OPNL TWI Access Error (SCF-8002-LX) for possible resolutions.

Details

I2C access failure to an I2C device on a FRU has occurred.

The I2C device on the FRU can no longer be accessed or controlled by the XSCF.

The I2C device can be identified from the FMRI of the ereport associated with the fault event.

This fault can occur on one of the many FRUs or devices supported on the SPARC Enterprise platforms.

The following section provides a list of the potential FRUs and devices
that could be impacted by this fault and specifies the impact and
automated actions taken by the platform or domain in response to the fault.



   SPARC Enterprise M3000/M4000/M5000 platforms:


      Extended System Controller Unit (XSCFU):


         The XSCFU on the M3000 platform is part of the Motherboard FRU.

         Fault due to an I2C access failure to an I2C device on the XSCFU.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the XSCFU consist of sensors (LED, voltage),
         controllers (voltage, 4-channel I2C MUXes), and a FRUID SEEPROM.
 
            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the platform cannot be restarted after a platform reset or poweron.


      Operator Panel (OPNL):

         Fault due to an I2C access failure to an I2C device on the Operator Panel (OPNL).
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the Operator Panel consist of sensors (temperature, LED),
         controllers (keyswitches, 4-channel I2C MUXes), SEEPROMs, and FRUID SEEPROM.
   
            If the resource that is inaccessible is one of the temperature sensors,
            then all the fans on the entire platform are turned to high speed
            and nothing is deconfigured.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the OPNL is deconfigured the next time the platform is restarted or powered on.


      Power Supply Unit (PSU):

         Fault due to an I2C access failure to an I2C device on a Power Supply Unit (PSU).
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the PSU consist of sensors (LED), controllers (fan), and a FRUID SEEPROM.

            If the resource that is inaccessible is a sensor,
            then all the fans on the entire platform are raised to high speed.
            Nothing is deconfigured and no immediate action taken.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the PSU is deconfigured the next time the platform is restarted or powered on.


      Fan Backplane (FANBP):

         Fault due to an I2C access failure to an I2C device on a Fan Backplane.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the Fan Backplane consist of sensors (LED, voltage, fan rotation),
         controllers (voltage, fan, fan hotplug controllers, 4-channel I2C MUXes),
         and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured and no immediate action is taken.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the platform cannot be restarted after a platform reset or poweron.

   
      DDC Riser (DDCR):

         Fault due to an I2C access failure to an I2C device on a DDC Riser.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C device on the DDC Riser consist of controllers (voltage) and a FRUID SEEPROM.

            Nothing is deconfigured and no immediate action is taken.
            The platform cannot be restarted after a platform reset or poweron.


      IO Unit (IOU):

         Fault due to an I2C access failure to an I2C device on an IOU.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on an IOU consist of IOC chips, FLP chips, sensors (temperature, LED, voltage),
         controllers (SAS repeaters, PCI-X Bridges, voltage, power, 4-channel I2C MUXes),
         and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured..

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the platform cannot be restarted after a platform reset or poweron.

            If the resource that is inaccessible is an IOC chip (IOC),
            then the IOC is deconfigured (affecting possibly two domains) after
            the domains using the IOC reboot or stop using the IOC due to a DR operation.

            
               Availability of I/O devices depends on which IOC Chip was deconfigured.


               SPARC Enterprise M3000 platforms:

                 There is no IOU on M3000 platform. 
                  The Oberon (IOC) chip and the FLP chip reside on the Motherboard.
                  The platform becomes unbootable.



               SPARC Enterprise M4000 platforms:

                  All IO on the platform is deconfigured and the platform becomes unbootable.

     

               SPARC Enterprise M5000 platforms:

                  IOC chip 0 provides access to PCI-X slot 0, PCI-Express slots 1-4,
                  and all the internal I/O.

                     IOU#0: Internal I/O includes the DVD drive, the DAT drive, HDD#0 and HDD#1.
                     IOU#1: Internal I/O includes HDD#2 and HDD#3.


 
            If the resource that is inaccessible is an FLP,
            then the IO Channel associated with the FLP chip is deconfigured when the domain
            using the IO Channel reboots or stops using the IO Channel due to a DR operation.


               Availability of I/O devices depends on which IO Channel was deconfigured.


               SPARC Enterprise M3000 platforms:

                 The platform becomes unbootable.



               SPARC Enterprise M4000 platforms:
   
                  IO Channel 0 provides access to PCI-X slot 0, PCI-Express slots 1 and 2,
                  and all the internal I/O.

                  IO Channel 1 provides access to PCI-Express slots 3 and 4.



               SPARC Enterprise M5000 platforms:

                  IO Channel 0 provides access to PCI-X slot 0, PCI-Express slots 1 and 2,
                  and all the internal I/O.

                  IOU#0: Internal I/O includes the DVD drive, the DAT drive, HDD#0 and HDD#1.
                  IOU#1: Internal I/O includes HDD#2 and HDD#3.

                  IO Channel 1 provides access to PCI-Express slots 3 and 4.


      DIMM (MEM):

         Fault due to an I2C access failure to the SPD of a DIMM.
         The SPD can no longer be accessed or controlled by the XSCF.

            No immediate action is taken.
            The DIMM is deconfigured once the domain using the DIMM reboots
            or the domain stops using the DIMM due to a DR operation.
            This means that an entire bank of memory (either 4 DIMMs or 8 DIMMs,
            depending upon the configuration) is deconfigured.


      Memory Board (MEMB):

         Fault due to an I2C access failure to an I2C device on a Memory Board.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a Memory Board consist of MAC chips, sensors (voltage),
         controllers (voltage, power, 4-channel I2C MUXes), and a FRUID SEEPROM.

        
          SPARC Enterprise M3000 platforms:

            There is no MEMB on the M3000 platform.
            The MAC chip, which is really the JSC chip,  is the on the Motherboard.
            The platform becomes unbootable.


           SPARC Enterprise M4000/M5000 platforms:

             If the resource that is inaccessible is one of the sensors,
             then nothing is deconfigured.

             If the resource that is inaccessible is either the MAC chip, one of the
             controllers, or the seeprom, then the Memory Board is deconfigured
             after the domain using the Memory Board reboots or
             the domains stop using the Memory Board due to a DR operation.


      CPU Module (CPUM):

         Fault due to an I2C access failure to an I2C device on a CPU Module.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a CPU Module consist of sensors (temperature, voltage),
         controllers (voltage, power, 4-channel I2C MUXes), and a FRUID SEEPROM.

            SPARC Enterprise M3000 platforms:

                There is no CPUM on Ikkaku systems.
                The CPU chip is the on the Motherboard.


            If the resource that is inaccessible is one of the sensors,
            then all the fans on the entire platform are turned to high speed
            and nothing is deconfigured.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the CPU Module is deconfigured once the domain using the CPU Module has
            rebooted or the domains have stopped using the CPU Module due to a DR operation.


      Motherboard (MBU):

         Fault due to an I2C access failure to an I2C device on a Motherboard.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a Motherboard Board consist of sensors (temperature, LED, voltage),
         controllers (voltage, 4-channel I2C MUXes), and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then all the fans on the entire platform are turned to high speed
            and nothing is deconfigured.

            If the resource that is inaccessible is one of the controllers, or the seeprom,
            then the platform cannot be restarted after a platform reset or poweron.




   SPARC Enterprise M8000/M9000 platforms:


      Extended System Controller Unit (XSCFU):

         Fault due to an I2C access failure to an I2C device on the XSCFU.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the XSCFU consist of sensors (LED, voltage),
         controllers (voltage, 4-channel I2C MUXes), and a FRUID SEEPROM.
 
            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the XSCFU is deconfigured the next time the XSCFU is restarted.

               If this is the active XSCFU,
               then the standby XSCFU will take over as the new active XSCFU.


      Extended System Controller Unit (XSCFU_C):
 
         Fault due to an I2C access failure to an I2C device on the XSCFU_C (XSCFU repeater).
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the XSCFU consist of sensors (LED, voltage),
         controllers (voltage), and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the partner XSCFU to this XSCFU_C is deconfigured the next time the
            XSCFU is restarted.

            If this is the active XSCFU,
            then the standby XSCFU will take over as the new active XSCFU.


      Operator Panel (OPNL):

         Fault due to an I2C access failure to an I2C device on the Operator Panel (OPNL).
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the Operator Panel consist of sensors (LED),
         controllers (keyswitches, 4-channel I2C MUXes), SEEPROMs, and FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured and no immediate action is taken.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the OPNL is deconfigured the next time the platform is restarted or powered on.


      Power Supply Unit (PSU):

         Fault due to an I2C access failure to an I2C device on a Power Supply Unit (PSU).
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the PSU consist of sensors (LED), controllers (fan), and a FRUID SEEPROM.

            If the resource that is inaccessible is a sensor,
            then all the fans on the entire platform are raised to high speed.
            Nothing is deconfigured and no immediate action taken.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the PSU is deconfigured the next time the platform is restarted or powered on.


      Sensor Unit (SNSU):

         Fault due to an I2C access failure to an I2C device on the Sensor Unit (SNSU).
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the Sensor Unit (SNSU) consist of temperature and humidity sensors.

            Since the resource that is inaccessible may be a temperature sensor,
            then all the fans on the entire platform are turned to high speed
            and nothing is deconfigured.


      FAN Backplane (FANBP):

         Fault due to an I2C access failure to an I2C device on a Fan Backplane.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the Fan Backplane consist of sensors (LED, fan rotation),
         controllers (voltage, fan, 4-channel I2C MUXes), and a FRUID SEEPROM.
 
            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured and no immediate action is taken.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the fan backplane is deconfigured the next time the platform is restarted
            or powered on.


      FAN:

         Fault due to an I2C access failure to an I2C device on a Fan Tray.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C device on the Fan Tray consists of a FRUID SEEPROM.

            Nothing is deconfigured and no immediate action is taken.


      IO Unit (IOU):

         Fault due to an I2C access failure to an I2C device on an IOU.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on an IOU consist of IOC chips, FLP chips, sensors (temperature, LED, voltage),
         controllers (SAS repeaters, PCI-X Bridges, voltage, power, 4-channel I2C MUXes),
         and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured..

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the IOU is deconfigured the after the domains using the IOU reboot or
            after the domains stop using the IOU due to a DR operation.

            If the resource that is inaccessible is an IOC chip (IOC),
            then the IOC is deconfigured (affecting possibly two domains) when
            the domains using the IOC reboot or stop using the IOC due to a DR operation.


               Availability of I/O devices depends on which IOC Chip was deconfigured.


               SPARC Enterprise M8000/M9000 platforms:

   
                  IOC chip 0 provides access to PCI-Express slots 0 and 1 on this IOU.

                  IOC chip 0 provides access to PCI-Express slots 2 and 3 on this IOU.

                  IOC chip 1 provides access to PCI-Express slots 4 and 5 on this IOU.

                  IOC chip 1 provides access to PCI-Express slots 6 and 7 on this IOU.

 

            If the resource that is inaccessible is an FLP,
            then the IO Channel associated with the FLP chip is deconfigured when the domain
            using the IO Channel reboots or stops using the IO Channel due to a DR operation.


               Availability of I/O devices depends on which IO Channel was deconfigured.


               SPARC Enterprise M8000/M9000 platforms:

   
                  IO Channel 0 on IOC chip 0 provides access to PCI-Express slots 0 and 1 on this IOU.

                  IO Channel 1 on IOC chip 0 provides access to PCI-Express slots 2 and 3 on this IOU.

                  IO Channel 0 on IOC chip 1 provides access to PCI-Express slots 4 and 5 on this IOU.

                  IO Channel 1 on IOC chip 1 provides access to PCI-Express slots 6 and 7 on this IOU.


      DIMM (MEM):

         Fault due to an I2C access failure to the SPD of a DIMM.
         The SPD can no longer be accessed or controlled by the XSCF.

            No immediate action is taken.
            The DIMM is deconfigured once the domain using the DIMM reboots
            or the domain stops using the DIMM due to a DR operation.
            This means that an entire bank of memory (either 4 DIMMs or 8 DIMMs,
            depending upon the configuration) is deconfigured.


      CPU Module (CPUM):

         Fault due to an I2C access failure to an I2C device on a CPU Module.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a CPU Module consist of sensors (temperature, voltage),
         controllers (voltage, power, 4-channel I2C MUXes), and a FRUID SEEPROM.
   
            If the resource that is inaccessible is one of the sensors,
            then all the fans on the entire platform are turned to high speed
            and nothing is deconfigured.

            If the resource that is inaccessible is one of the controllers or the seeprom,
            then the CPU Module is deconfigured once the domain using the CPU Module has
            rebooted or the domains have stopped using the CPU Module due to a DR operation.


      CPU/Memory Unit (CMU):


         Fault due to an I2C access failure to an I2C device on a CMU.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a CMU consist of MAC chips, sensors (temperature, LED, voltage),
         controllers (voltage, 4-channel I2C MUXes), and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then all the fans on the entire platform are turned to high speed
            and nothing is deconfigured.

            If the resource that is inaccessible is a MAC chip,
            then the MAC chip is deconfigured, along with the MAC chip this
            MAC chip is striping with, after the domains using the MAC chips
            reboot or the domains stop using the MAC chips due to a DR operation.
       
            If the resource that is inaccessible is either one of the controllers
            or the seeprom, then the CMU is deconfigured once the domains using the
            CMU have rebooted or the domains have stopped using the CMU due to a DR operation.


      DDC_A:

         Fault due to an I2C access failure to an I2C device on the DDC_A.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C device on the DDC_A consist of a FRUID SEEPROM.

            If the resource that is inaccessible is the seeprom, then the DDC_A is deconfigured
            the next time the platform is restarted or powered up.

            If the other DDC_A is non-operational, then the platform can't be restarted.
         

      Backplane (BP_A):

         Fault due to an I2C access failure to an I2C device on the Backplane BP_A.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on the BP_A consist of sensors (temperature, LED, voltage),
         controllers (voltage), and a FRUID SEEPROM.
    
            If the resource that is inaccessible is one of the sensors,
            then all the fans on the entire platform are turned to high speed
            and nothing is deconfigured and no immediate action is taken.

            If the resource that is inaccessible is one of the controllers, or the seeprom,
            then the domains will continue to run, but the platform can't be restarted or powered up.


      Crossbar Unit (XBU):

         Fault due to an I2C access failure to an I2C device on the Crossbar Unit (XBU).
         These devices can no longer be accessed or controlled by the XSCF.
 
         I2C devices on an XBU consist of sensors (temperature, LED, voltage),
         controllers (voltage), and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then all the fans on the entire platform are turned to high speed
            and nothing is deconfigured.

            If the resource that is inaccessible is either one of the controllers
            or the seeprom, then the Crossbar way is deconfigured the next time
            the platform is restarted or powered on.


      Clock Unit (CLKU):

         Fault due to an I2C access failure to an I2C device on the Clock Unit.
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a Clock Unit consist of sensors (temperature, LED, voltage),
         controllers (voltage), and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured and no immediate action is taken.

            If the resource that is inaccessible is either one of the controllers
            or the seeprom, then the Clock Unit is deconfigured the next time the
            platform is restarted or powered on.


      Switch Backplane (SWBP):
      
         Fault due to an I2C access failure to an I2C device on a Switch Backplane (SWBP)
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a Switch Backplane consist of SAS Expanders, sensors (voltage),
         controllers (voltage, 4-channel I2C MUXes), and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured and no immediate action is taken.

            If the resource that is inaccessible is the one of the controllers or
            the SEEPROM, then the SWBP will be deconfigured once the domain
            using the DVD/Tape reboots or the domain stops using the DVD/Tape.
            (The DVD and Tape Unit will be inaccessible)


      Media Backplane (MEDBP):

         Fault due to an I2C access failure to an I2C device on a Media Backplane (MEDBP)
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a Media Backplane consist of sensors (voltage) and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured and no immediate action is taken.

            If the resource that is inaccessible is the SEEPROM,
            then the SWBP will be deconfigured once the domain
            using the DVD/Tape reboots or the domain stops using the DVD/Tape.
            (The DVD and Tape Unit will be inaccessible).


      DVD Unit (DVDU):

         Fault due to an I2C access failure to an I2C device on a DVD Unit (DVDU)
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a DVD Unit consist of sensors (voltage) and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured and no immediate action is taken.

            If the resource that is inaccessible is the SEEPROM,
            then the SWBP will be deconfigured once the domain
            using the DVD/Tape reboots or the domain stops using the DVD/Tape.
            (The DVD and Tape Unit will be inaccessible).


      Tape Unit (TAPEU):

         Fault due to an I2C access failure to an I2C device on a Tape Unit (TAPEU).
         These devices can no longer be accessed or controlled by the XSCF.

         I2C devices on a Tape Unit consist of sensors (voltage) and a FRUID SEEPROM.

            If the resource that is inaccessible is one of the sensors,
            then nothing is deconfigured and no immediate action is taken.

            If the resource that is inaccessible is the SEEPROM,
            then the SWBP will be deconfigured once the domain
            using the DVD/Tape reboots or the domain stops using the DVD/Tape.
            (The DVD and Tape Unit will be inaccessible).


The recommended service action for this event is to schedule the replacement of the affected FRU's.


Step 1. Collect the fault message (use one of the following methods):


   Single-line fault message displayed on the XSCF console:

   Mar 27 15:06:08 nwk-dc2-1-sc0 fmd: SOURCE: sde, REV: 1.12, CSN: nwk-dc2-1  
   EVENT-ID: 60dc91be-23bf-4770-805c-fa66f72fae34
   Refer to http://www.sun.com/msg/ SCF-8002-LX for detailed information.


   Complete fault message using 'fmdump -m' on the XSCF console:

   MSG-ID:  SCF-8002-LX, TYPE: Fault, VER: 1, SEVERITY: Critical
   EVENT-TIME: Tue Mar 27 15:06:07 PDT 2007
   PLATFORM: SPARC-Enterprise, CSN: nwk-dc2-1, HOSTNAME: nwk-dc2-1-sc0
   SOURCE: sde, REV: 1.12
   EVENT-ID: 60dc91be-23bf-4770-805c-fa66f72fae34
   DESC: I2C access failure to an I2C device.
   Refer to http://www.sun.com/msg/SCF-8002-LX for more information.
   AUTO-RESPONSE: The resources associated with the I2C device may deconfigured.
   The speed of all fans on the platform may be raised.
   Please consult the detail section of the knowledge article for additional information.
   IMPACT: It is possible that the platform will not be able to be be restarted.
   Please consult the detail section of the knowledge article for additional information.
   REC-ACTION: The platform administrator should schedule a service action to replace
   the affected Field Replaceable Unit (FRU), the identity of which can be
   determined using fmdump -v -u EVENT_ID.
   Please consult the detail section of the knowledge article for additional information.


Step 2. Collect the output from the 'fmdump -v -u EVENT_ID' command
 

   SPARC Enterprise platform example:

   xscf> fmdump -v -u 60dc91be-23bf-4770-805c-fa66f72fae34

         TIME                 UUID                                 MSG-ID
         Mar 27 15:06:07.9710 60dc91be-23bf-4770-805c-fa66f72fae34 SCF-8002-LX
           66%  fault.chassis.i2c.device

                Problem in: hc:///chassis=0/swbp=0
                   Affects: hc:///chassis=0/swbp=0
                       FRU: hc://:product-id=SPARC-Enterprise:chassis-id=nwk-dc2-1:
                            server-id=nwk-dc2-1-sc0:serial=PP06143682:
                            part=CA20394-B16X 006AE:revision=0101/component=/SWBP#0

           33%  fault.chassis.i2c.device

                Problem in: hc:///chassis=0/medbp=0
                   Affects: hc:///chassis=0/swbp=0
                       FRU: hc://:product-id=SPARC-Enterprise:chassis-id=nwk-dc2-1:
                            server-id=nwk-dc2-1-sc0:serial=PP0639P343:
                            part=CA20396-B66X 005AC:revision=0101/component=/MEDBP#0

Step 3. Collect a snapshot ( See document 2097446.1 - SPARC Mx000 and M10/M12 systems: Simple Instructions to Collect an XCP Snapshot).

Step 4.  Contact your Authorized Service Provider.

If you require additional information, please refer to Document 1002526.1.