ASM启动ORA-27125错误诊断-CSDN博客

帮前同事诊断一个ASM无法启动的问题。

数据库环境Oracle 10.2.0.3 RAC for Solaris 10 sparc64，由于操作系统故障，导致实例2主机无法启动，在尝试重做系统时发现实例1上的ASM无法正常启动。

详细错误信息为：

bash-3.00# su - oracle
Oracle Corporation SunOS 5.10 Generic Patch January 2005
bash-3.00$ export ORACLE_SID=+ASM1
bash-3.00$ sqlplus / as sysdba

SQL*Plus: Release 10.2.0.3.0 - Production on Sun Jun 5 20:14:01 2011

Connected to an idle instance.

SQL> startup
ORA-27125: unable to create shared memory segment
SVR4 Error: 14: Bad address
SQL> exit
Disconnected

这个ORA-27125的错误信息到是很常见，一般来说都是/etc/system中参数配置不合理导致SGA分配受到限制。

不过对于ASM实例而言，分配的SGA很有限，不应该出现这个错误，而且随后的SVR4 Error: 14: Bad address错误信息更是说明的错误并不一般。

打算通过ipcs检查当前是否已经存在共享内存段，结果发现了错误：

bash-3.00$ ipcs
IPC status from as of Sun Jun 5 20:14:43 CST 2011
Bad System Call (core dumped)

第一次运行出现Bad System Call (core dumped)错误，而再次运行还可能碰到下面的错误：

# ipcs
IPC status from as of Sun Jun 5 20:57:20 CST 2011
T ID KEY MODE OWNER GROUP
Message Queues:
shmids: Bad address

检查了/etc/system文件，虽然有些配置不太合理，但是理论上讲不应该出现这个错误。修改这个文件，并根据Oracle推荐的值进行了修改，重新启动主机，问题依旧。

检查操作系统日志，果然发现了错误：

Jun 5 20:49:23 ser02 genunix: [ID 936769 kern.info] emcp8 is /pseudo/emcp@8
Jun 5 20:49:23 ser02 pseudo: [ID 129642 kern.info] pseudo-device: emcp14
Jun 5 20:49:23 ser02 genunix: [ID 936769 kern.info] emcp14 is /pseudo/emcp@14
Jun 5 20:49:23 ser02 scsi: [ID 107833 kern.warning] WARNING: /pci@8,600000/QLGC,qlc@1/fp@0,0/ssd@w5006016341e0ac1c,0 (ssd1):
Jun 5 20:49:23 ser02 drive offline
Jun 5 20:49:23 ser02 scsi: [ID 107833 kern.warning] WARNING: /pci@8,600000/QLGC,qlc@1/fp@0,0/ssd@w5006016341e0ac1c,0 (ssd1):
Jun 5 20:49:23 ser02 drive offline
Jun 5 20:49:23 ser02 scsi: [ID 107833 kern.warning] WARNING: /pci@8,600000/fibre-channel@2/fp@0,0/ssd@w5006016241e0ac1c,0 (ssd8):
Jun 5 20:49:23 ser02 drive offline
Jun 5 20:49:23 ser02 scsi: [ID 107833 kern.warning] WARNING: /pci@8,600000/fibre-channel@2/fp@0,0/ssd@w5006016241e0ac1c,0 (ssd8):
Jun 5 20:49:23 ser02 drive offline
Jun 5 20:49:23 ser02 scsi: [ID 107833 kern.warning] WARNING: /pci@8,600000/fibre-channel@2/fp@0,0/ssd@w5006016a41e0ac1c,3 (ssd10):
Jun 5 20:49:23 ser02 drive offline
Jun 5 20:49:23 ser02 scsi: [ID 107833 kern.warning] WARNING: /pci@8,600000/fibre-channel@2/fp@0,0/ssd@w5006016a41e0ac1c,3 (ssd10):
Jun 5 20:49:23 ser02 drive offline
Jun 5 20:50:12 ser02 genunix: [ID 482702 kern.notice] sys/shmsys:shmexit() not defined properly
Jun 5 20:58:42 ser02 snmpXdmid: [ID 216524 daemon.error] Registration with DMI failed. err = 831.