今天查看两个新装的库的进程，发现有很多dbsnmp相关的进程。

oracle 20368 20365 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20369 20368 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20370 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20374 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20375 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20376 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20377 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20378 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20379 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20380 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20381 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20382 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20383 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20384 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20385 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20386 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp
oracle 20387 20369 0 Apr05 pts/3 00:00:00 /opt/oracle/product/9.2/bin/dbsnmp

重启数据库问题依旧，查了一下dbsnmp是跟智能代理相关的一个命令，尝试手工执行：
$ /opt/oracle/product/9.2/bin/dbsnmp
DBSNMP for Linux 2.6.9-5.ELsmp: release 9.2.0.4.0 - Production on 09-APR-07 09:15:47
Discovering services on datasrv2 configured on host addresses {datasrv2 localhost 127.0.0.1}
Using oratab file /etc/oratab
Warning : LISTENER defined in /opt/oracle/product/9.2/network/admin/listener.ora will be skipped, because it does not monitor any of the valid SIDs {testcen}
Warning : No Listener found for SID testcen. testcen will be skipped

看报错信息似乎与监听没有配置SID相关，查看listener.ora如下：
SID_LIST_LISTENER =
(SID_LIST =
(SID_DESC =
(SID_NAME = PLSExtProc)
(ORACLE_HOME = /opt/oracle/product/9.2)
(PROGRAM = extproc)
)
)

LISTENER =
(DESCRIPTION_LIST =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = datasrv2)(PORT = 1521))
)
(DESCRIPTION =
(ADDRESS = (PROTOCOL = IPC)(KEY = EXTPROC))
)
)

往监听文件上添加上sid相关信息
SID_LIST_LISTENER =
(SID_LIST =
(SID_DESC =
(SID_NAME = PLSExtProc)
(ORACLE_HOME = /opt/oracle/product/9.2)
(PROGRAM = extproc)
)
(SID_DESC =
(GLOBAL_DBNAME = testcen)
(ORACLE_HOME = /opt/oracle/product/9.2)
(SID_NAME = testcen)
)
)

LISTENER =
(DESCRIPTION_LIST =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = datasrv2)(PORT = 1521))
)
(DESCRIPTION =
(ADDRESS = (PROTOCOL = IPC)(KEY = EXTPROC))
)
)

然后重启库，再执行：
[oracle@datasrv2 admin]$ ps -ef|grep dbsnmp
oracle 3559 3496 0 09:45 pts/2 00:00:00 grep dbsnmp

可以看到问题已经解决。问题出现原因估计是因为oracle在不能启动dbsnmp时会不断重试，从而导致有多个dbsnmp的情况。

ITPub博客

dbsnmp启动失败解决方法

最新文章