X ORA-27303: additional information: startup egid = 202 (?), current egid = 101 (dba)

In this Document

Symptoms

Changes

Cause

Solution

---

APPLIES TO:

Oracle Database - Enterprise Edition - Version 10.2.0.3 to 11.2.0.3 [Release 10.2 to 11.2]
Oracle Database - Enterprise Edition - Version 12.1.0.2 to 12.1.0.2 [Release 12.1]
Information in this document applies to any platform.

SYMPTOMS

Connect " /as sysdba" is showing connected but failing when performing any query.

$ sqlplus "/as sysdba"

SQL*Plus: Release 10.2.0.3.0 - Production on Wed Aug 8 19:34:19 2007

Copyright (c) 1982, 2006, Oracle. All Rights Reserved.

Connected.
SQL> select * from v$instance;
select * from v$instance
*
ERROR at line 1:
ORA-01012: not logged on

CHANGES

gid of the $ORACLE_HOME/bin/oracle was changed when the database was running.

CAUSE

At the time of startup the background processes are having a gid, and since the gid of the
$ORACLE_HOME/bin/oracle was changed, the new gid was not matching with the gid at startup.

Below entries can be found in the trace file, ORA-27303 indicates the startup and current gid. 

Oracle Database 10g Enterprise Edition Release 10.2.0.3.0 - 64bit Production 
With the Partitioning, OLAP and Data Mining options 
ORACLE_HOME = <ORACLE_HOME> 
System name: AIX 
Node name: xxx1 
Release: 3 
Version: 5 
Machine: <MACHINE>
Instance name: <INSTANCE_NAME>
Redo thread mounted by this instance: 1 
Oracle process number: 0 
Unix process pid: 524430, image: oracle@xxx1 

Died during process startup with error 27140 (seq=805) 
OPIRIP: Uncaught error 27140. Error stack: 
ORA-27140: attach to post/wait facility failed 
ORA-27300: OS system dependent operation:invalid_egid failed with status: 1 
ORA-27301: OS failure message: Not owner 
ORA-27302: failure occurred at: skgpwinit6 
ORA-27303: additional information: startup egid = 202 (?), current egid = 101 (dba) 
oracle@xxx0: ABC : /opt/oracle/ABC/bdump

SOLUTION

Change the gid of the $ORACLE_HOME/bin/oracle to the original value i.e. startup gid.

=======================================================================================================

案例：

环境：
OS：oracle linux5.6
DB: oracle11gR2 PS3
RAC node:rac1,rac2

现象：
在创建数据库时，在复制数据文件时，就会出现
ora-19504:failed to create file “+DATA”
……
ora-15055:unable to connect ASM instance
ora-27303:additional information:startup egid=54321(oinstall) ,current egid=54324(dba)
……

 

解决：
从报错信息可以看出，可能是asm磁盘组权限问题。
Db software安装完成后，$ORACLE_HOME/bin/oracle文件属性如下：
[oracle@rac1 ~]$ cd $ORACLE_HOME/bin/
[oracle@rac1 bin]$ chown oracle:oinstall oracle
[oracle@rac1 bin]$ ls -l oracle
-rwxr-x--x 1 oracle oinstall 232399431 10-20 14:13 oracle

在出现上面创建库报错信息后，再查看此文件
[oracle@rac2 dev]$ cd $ORACLE_HOME/bin
[oracle@rac1 bin]$ ls -l oracle
-rwxr-s--x 1 oracle dba 232399431 10-20 14:13 oracle

这样发现在dbca的过程中$ORACLE_HOME/bin/oracle文件属性会自动被修改，可能就是因为这个原因，导致在属性修改后对ASM组没有了足够的权限创建数据文件。

然后再回顾前面GI的安装过程，在选择OSASM组时选择的dba属组，应该是这里asm属组配置的问题。如下：

分析了问题原因后，重新部署安装GI，在OSASM组选择oinstall安装，如下：

配置安装完成后，在创建数据库时不会再报错。

再查看$ORACLE_HOME/bin/oracle文件属性
[oracle@rac2 dev]$ cd $ORACLE_HOME/bin
-rwxr-s--x 1 oracle oinstall 232399431 10-20 14:13 oracle

注：以后再遇到这种问题，很可能就是安装的过程中，没有注意用户权限的选择。可以和正常安装系统对比一下文件权限。

如$ORACLE_HOME/bin/oracle,$GRID_HOME/bin/oracle文件权限是否正确，修改后再试。

另外，也可以参考这篇文章解决：

http://blog.csdn.net/konglyon/article/details/7322590

参考文档：
Matalink文档：
ID 453959.1 
ID 1054033.1

杨大师：
http://yangtingkun.itpub.net/post/468/523429