Oracle内部错误:ORA-00600[OSDEP_INTERNAL]一例

一套HP-UX上的9.2.0.5系统在shutdown abort时出现ORA-00600: internal error code, arguments: [OSDEP_INTERNAL], [], [], [], [], [], [], []内部错误，伴随有ORA-27302: failure occurred at: skgpwinit4，ORA-27303: additional information: attach to invalid skgp shared ctx，具体日志如下:

/opt/oracle/product/9.2.0.5/rdbms/log/ngende_ora_7669.trc
Oracle9i Enterprise Edition Release 9.2.0.5.0 - 64bit Production
With the Partitioning, OLAP and Oracle Data Mining options
JServer Release 9.2.0.5.0 - Production
ORACLE_HOME = /opt/oracle/product/9.2.0.5
System name: HP-UX
Node name: yictngd3
Release: B.11.23
Version: U
Machine: ia64
Instance name: nGende
Redo thread mounted by this instance: 0 
  
   
Oracle process number: 0
7669

*** 2010-09-08 00:10:02.985
ksedmp: internal or fatal error
ORA-00600: internal error code, arguments: [OSDEP_INTERNAL], [], [], [], [], [], [], []
ORA-27302: failure occurred at: skgpwinit4
ORA-27303: additional information: attach to invalid skgp shared ctx
Current SQL information unavailable - no session.

Call stack
--------------
ksedmp <- ksfdmp <- kgerinv <- kgerin <- kgerecoserr <- ksucrp <- ksucresg <- kpolna
<- kpogsk <- opiodr <- ttcpip <- opitsk <- Cannot <- Cannot <- Cannot <- Cannot <- opiino
<- opiodr <- opidrv <- sou2o <- main <- main_opd_entry

  

经查该内部错误与操作系统共享内存有关，相关的Note有:

Ora-00600: Internal Error Code, Arguments: [Osdep_internal] [ID 304027.1]
Applies to:
Oracle Server - Enterprise Edition - Version: 9.2.0.2 to 10.2.0.3 - Release: 9.2 to 10.2
Information in this document applies to any platform.
***Checked for relevance on 03-NOV-2010***

Getting ORA-600 [OSDEP_INTERNAL] errors while starting up the database:

ORA-00600: internal error code, arguments: [OSDEP_INTERNAL],
[], [], [], [], [], [], []
ORA-27302: failure occurred at: skgpwreset1
ORA-27303: additional information: invalid shared ctx
ORA-27146: post/wait initialization failed
ORA-27300: OS system dependent operation:semget failed with status: 28
ORA-27301: OS failure message: No space left on device
ORA-27302: failure occurred at: sskgpsemsper
Symptoms
Getting ORA-600 [OSDEP_INTERNAL]
Accompanied by the following errors
ORA-27302:Failure occured at: skgpwreset1
ORA-27303:additional information: invalid shared ctx
ORA-27146: post/wait initialization failed
ORA-27300: OS system dependent operation: segment failed with error 28
ORA-27301: OS system Failure message: No space left on device
ORA-27302: failure occured at: sskgpsemsper

Cause
The functions in the trace file generated point to the semaphore settings .
Smmns is set too low.

Solution
set semmns 32767
Arrange to make the changes persistent as per the Operating system then restart the server and check if the changes are persistent.
eg: Linux /etc/sysctl.conf

sem = semmsl semmns semopm semmni
kernel.sem = 256 32768 100 228

Getting ORA-00600 [OSDEP_INTERNAL]: Internal Error While Trying To Connect / As Sysdba [ID 253885.1]

Applies to:
Oracle Server - Enterprise Edition - Version: 9.2.0.3 and later   [Release: 9.2 and later ]
HP-UX PA-RISC (64-bit)
Symptoms
Getting following error while trying to connect as sysdba using sqlplus:

SQL> conn / as sysdba
ERROR:
ORA-01041: internal error. hostdef extension doesn't exist

Alert.log shows:

ORA-00600: internal error code, arguments: [OSDEP_INTERNAL], [], [], [], [], [],[], []
ORA-27302: failure occurred at: skgpwinit4
ORA-27303: additional information: attach to invalid skgp shared ctx
Cause
- Database was shutdown using "shutdown abort" option.
- Shared memory segment was not removed even though the instance was down.
Solution
+ Check which shared memory segments are owned by the oracle owner

Use the ipcs -bm command:

% ipcs -bm

m 34034336 0xf8f18468 --rw-r----- ORACLE dba 16777216

+ Delete the 'orphan' shared memory segments:

% ipcrm -m 34034336

If there is more than one instance running on the server and you are not sure how to identify the shared
memory segments then please contact support.

不恰当的设置OS VM参数可能导致该问题，而在HP-UX PA-RISC平台上使用'shotdown abort'命令时可能因为共享内存未能正常移除而出现该内部错误；因为实例还是以'abort'方式关闭的，仅仅是共享内存未能释放，所以只需要以ipcs->ipcrm等os命令将相应的共享内存段释放就可以了，不会造成其他影响。