http://space.itpub.net/540962/viewspace-708279
有一次，装完数据库后，并配置好监听，当使用SQL*PLUS,PL/SQL DEVELOPER工具登录都没有问题，但是使用OEM管理界面的时候就报错了，下面是我的整个排错过程，
1）检查告警日志（alert_SID.log)

仔细看了下日志，没有相关的报警信息，同时也查看了listener.ora的文件，连接也正常

2）检查监听文件

由于用其它工具可以连接，首先当时我没有怀疑监听的问题，我记得当时建监听文件是从正式库运行的拷贝过来的，启动完监听后，信息如下：

lsnrctl stat

LSNRCTL for Linux: Version 10.2.0.4.0 - Production on 06-DEC-2010 10:25:27

Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=10.10.100.99)(PORT=1521)))
STATUS of the LISTENER
22275400------------------------
Alias                     LISTENER
Version                   TNSLSNR for Linux: Version 10.2.0.4.0 - Production
Start Date                06-DEC-2010 10:18:17
Uptime                    0 days 0 hr. 7 min. 10 sec
Trace Level               off
Security                  ON: Local OS Authentication
SNMP                      OFF
Listener Parameter File   /u01/app/oracle/db10g/network/admin/listener.ora
Listener Log File         /u01/app/oracle/db10g/network/log/listener.log
istening Endpoints Summary...
b1W5`)N#? }a22275400 (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=df2)(PORT=1521)))
Services Summary...
Service "df2" has 2 instance(s).
Instance "df2", status UNKNOWN, has 1 handler(s) for this service...

当时就怀疑这个监听有点怪，后面经开发人员反映也没有什么问题，就一直没有去理会，经过二个星期，由于使用oracle 10g OEM的功能进行数据库的监听就出现了上述的问题

3）启用oem

emctl start dbconsole --正常

4）浏览http://10.10.100.99:1158/em/ 就出现了访问不了，提示是监听的问题

5）后面根据网上的一些方法，说是listener.ora和tnsnames.ora文件的配置问题，不过我仔细看了下IP,PORT,ORACLE_HOME，没有什么问题，苦苦想了很久，从报错的信息来看，还是监听的问题，然后使用netca重新进行了配置，发现用lsnrctl stat终于有了不一样的变化了，如下：

LSNRCTL for Linux: Version 10.2.0.4.0 - Production on 06-DEC-2010 10:25:27

Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=10.10.100.99)(PORT=1521)))
STATUS of the LISTENER
Version                   TNSLSNR for Linux: Version 10.2.0.4.0 - Production
Start Date                06-DEC-2010 10:18:17
Uptime                    0 days 0 hr. 7 min. 10 sec
Trace Level               off
Security                  ON: Local OS AuthenticationI
SNMP                      OFF
Listener Parameter File   /u01/app/oracle/db10g/network/admin/listener.ora
Listener Log File         /u01/app/oracle/db10g/network/log/listener.log
Listening Endpoints Summary...
(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=df2)(PORT=1521)))
Services Summary...
Service "df2" has 2 instance(s).
Instance "df2", status UNKNOWN, has 1 handler(s) for this service...
Instance "df2", status READY, has 1 handler(s) for this service...
Service "df2XDB" has 1 instance(s).
Instance "df2", status READY, has 1 handler(s) for this service...
Service "df2_XPT" has 1 instance(s).
Instance "df2", status READY, has 1 handler(s) for this service...
The command completed successfully

就像看到了久违的老朋友一样，让我欢喜万分，马上打开浏览器，输入地址一看，错误消失，监控浏览正常

总结：

导致这个问题的最终原因，是listener.ora配置的问题，后面进行了前后的listener.ora对比：

出错前：

# listener.ora Network Configuration File: /u01/app/oracle/db10g/network/admin/listener.ora
# Generated by Oracle configuration tools.

SID_LIST_LISTENER =
(SID_LIST =
   (SID_DESC =
      (GLOBAL_DBNAME = df2)
      (ORACLE_HOME = /u01/app/oracle/db10g)
      (SID_NAME = df2)
    )
)

LISTENER =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 10.10.100.99)(PORT = 1521))
)

解决后：

# listener.ora Network Configuration File: /u01/app/oracle/db10g/network/admin/listener.ora
# Generated by Oracle configuration tools.

SID_LIST_LISTENER =
(SID_LIST =
    (SID_DESC =
     (GLOBAL_DBNAME = df2)
   (ORACLE_HOME = /u01/app/oracle/db10g)
      (SID_NAME = df2)
    )
)

LISTENER =
(DESCRIPTION_LIST =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 10.10.100.99)(PORT = 1521))
)
)
标示红色的地方是有点不同，我后面想了下，由于之前是手工配置的文件，而且进行了修改，是标点符号误差原因造成，所以一定要谨慎监听文件的配置，为什么开发人员可以使用呢？看之前的监听状态，有一个是unkown的，意思是说监听是静态注册的，因此都可以使用，但OEM会调用相关的ORACLE内部的一些服务就最后失败了

ORA-12541: TNS:no listener (DBD ERROR: OCIServerAttach) (转)