My Oracle Support Banner

X5-4 (ZS5-4) Running Solaris Panics With a Clustron Heartbeat Failure Due to an SMI Failover and SPX86A-8002-F7 Alert (Doc ID 2749487.1)

Last updated on FEBRUARY 05, 2021

Applies to:

Oracle Server X5-4 - Version All Versions and later
Information in this document applies to any platform.

Symptoms

The system will panic due to a memory related issue.

Panic Stack:

intel.mc indicates the ereport originated from a memory controller. It is also mentioned in the body of the ereport (highlighted in the above).

 

You may also see an FMA alert with the event code SPX86A-8002-F7:

--------------- ------------------------------------  -------------- ---------
TIME            EVENT-ID                              MSG-ID         SEVERITY
--------------- ------------------------------------  -------------- ---------
Jan 21 16:52:25 <UUID>                         SPX86A-8002-F7 Minor

Problem Status    : open
Diag Engine       : fdd / 1.0
System
   Manufacturer  : Oracle Corporation
   Name          : ORACLE SERVER X5-4
   Part_Number   : 34679006+16+1
   Serial_Number : <Serial>
   Host_ID       : (null)
   Server_Name   : <Host name>

----------------------------------------
Suspect 1 of 2 :
  Fault class : fault.cpu.intel.mxb.smi2_half_width_failover
  Certainty   : 50%

  FRU
    Location         : "/SYS/MB/P1/MR1"
    Manufacturer     : unknown
    Name             : unknown
    Part_Number      : 7090451
    Revision         : 02
    Serial_Number    : <Serial>
    Chassis
       Manufacturer  : Oracle Corporation
       Name          : ORACLE SERVER X5-4
       Part_Number   : 34679006+16+1
       Serial_Number : <Serial>
       Status        : faulty
  Resource
    Location         : "/SYS/MB/P1/MR1/MXB1"
    Status           : faulted but still in service
----------------------------------------
Suspect 2 of 2 :
  Fault class : fault.cpu.intel.mxb.smi2_half_width_failover
  Certainty   : 50%

  FRU
    Location         : "/SYS/MB/P1"
    Manufacturer     : unknown
    Name             : unknown
    Part_Number      : CM80645
    Revision         :
    Serial_Number    : <Serial>
    Chassis
       Manufacturer  : Oracle Corporation
       Name          : ORACLE SERVER X5-4
       Part_Number   : 34679006+16+1
       Serial_Number : <Serial>
       Status        : faulty
  Resource
    Status           : faulted but still in service

Description : A Scalable Memory Interconnect (SMI) half-width failover has been
             detected.

Response    : SMI failover to half-width mode. The chassis wide and affected
             component service-required LEDs are illuminated.

Impact      : System continues to operate in the presence of this fault. SMI
             lanes are operating in half-width mode. System performance may be
             affected, as memory access time is reduced.

Action      : Please refer to the associated reference document at
             http://support.oracle.com/msg/SPX86A-8002-F7 for the latest
             service procedures and policies regarding this diagnosis.

 

Typically it will point at a processor and/or a riser.

 

There should also be additional ereports describing the fault that caused the above alert:

ereport.cpu.intel.mxb.smi2_half_width_failover@/SYS/MB/P1/MR1/MXB1

ereport.cpu.intel.mc_ce@/SYS/MB/P1

 

IPMI might also log errors (in this case memory ECC errors):

4b2 | 01/21/2021 | 16:52:13 | Memory #0x08 | Correctable ECC logging limit reached | Asserted | OEM Data-2 0x02 OEM Data-3 0x40
4c7 | 01/21/2021 | 17:42:03 | OEM HOST_ERR | State Asserted
4c8 | 01/21/2021 | 22:41:32 | Memory #0x08 | Correctable ECC logging limit reached | Asserted | OEM Data-2 0x02 OEM Data-3 0x40

Changes

 No changes are made, but ensure that the system has the latest firmware if possible.

Cause

To view full details, sign in with your My Oracle Support account.

Don't have a My Oracle Support account? Click to get started!


In this Document
Symptoms
Changes
Cause
Solution
References


My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.