Chassis code logging failure

passpark

Chassis code logging failure 的可能的原因分析。
Chassis code logging failure的原因可能有很多种，通过今天的case总结一下。
今天的情况是这样的：
Summary:
     Chassis code logging failure
Description of Error:
     An attempt by the operating system to log a chassis code failed.
Probable Cause / Recommended Action:
     Communications between the processor dependent code (PDC) and the service
     processor has failed.
          Verify service processor operation.  Enter chassis log manually with
          service processor commands.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^D E T A I L^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
An attempt by the operating system (OS) to do a chassis log write function
failed.
The chassis code that the OS was trying to write was:
   Encoded chassis code: 0x78e008041100f000
   Data field:           0x0000000200000000
The processor dependent code returned the status -5.  Possible status values:
   -1 = Unimplemented procedure.
   -2 = Nonexistent Option.
   -3 = Call completed with error.
  -10 = Invalid argument.
用CCV（Chassis Code Viewer）分析了一下 chassis code 0x78e008041100f000,结果如下：
Chassis code source is: Chassis Code Hex - Chassis Code Hex:  0x78e008041100f000
Field decode based on PDC_PAT ARS, version 2.6
Source[31:28]:           0x1( 1) - processor
  Source Detail[27:24]:  0x1( 1) - Processor General
  Source ID[23:16]:    0x00(  0)
  Event Detail[35:32]:   0x4( 4) - timeout
Reporting Entity Type[55:52]:   0xe(14) - HP-UX
  Reporting Entity ID[51:44]: 0x00(  0) - Cabinet 0, Cell 0, CPU 0
  Caller Activity[15:12]:       0xf(15) - display_activity() update
  Caller Subactivity[11:4]:   0x00(  0) - Implementation Dependent
  Activity Status[3:0]:         0x0( 0) - platform dependent
Data Type[63:59]:   0xf(15) - activity level / timeout
Data word description:
Alert Level[39:36]: 0x0( 0) - No failure detected, forward progress logs and timer requests only
***
Summary:
Description:
EOM[43:43]: 0x1(1) - No additional associated log entries follow.
Message ID[42:40]: 0x0(0)
Reserved[57:56]: 0x0(0)
关键字：timeout，所以错误的原因应该是timeout，如果改情况只是偶尔出现，应不属于很大的问题。一般通过run STM 或reset GSP (cm>XD ,r)解决。


本文来自ChinaUnix博客，如果查看原文请点：http://blog.chinaunix.net/u/3873/showart_11975.html