Oracle 11g RAC 监听器使用详解

182次阅读

共计 7815 个字符，预计需要花费 20 分钟才能阅读完成。

最近部署一个两节点 Oracle RAC 环境，刚刚开始没怎么留意，但当在使用的时候竟然 scan-ip 无法通过客户端登陆使用，但是在两个节点上可正常使用，此外 vip 无论在节点上还是其他客户端使用完全ok 。

在安装 GI 的时候最后报错，但经查看官方文档说是可能 scan-ip 写到 hosts 文件中了，但本人环境中三个 scan-ip 均为加入到 hosts 文件中，故此直接忽略掉该报错。

下面是本人详细的排错过程：

1. 在任意节点上使用 tnsnames.ORA 透过 scan-ip 登陆数据库：

[grid@RAC01 admin]$ tnsping RACDB

TNS Ping Utility for Linux: Version 11.2.0.1.0 – Production on 02-JAN-2014 18:41:16

Used parameter files:

/u/app/11.2.0/grid/network/admin/sqlnet.ora

Used TNSNAMES adapter to resolve the alias

Attempting to contact (DESCRIPTION = (ADDRESS_LIST = (ADDRESS = (PROTOCOL = TCP)(HOST = 10.134.30.50)(PORT = 1521))) (CONNECT_DATA = (SERVICE_NAME = RACDB)))

OK (0 msec)

[grid@RAC01 admin]$ sqlplus scott/testpassword@RACDB

SQL*Plus: Release 11.2.0.1.0 Production on Thu Jan 2 18:41:27 2014

Connected to:

Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 – 64bit Production

With the Partitioning, Real Application Clusters, Automatic Storage Management, OLAP,

Data Mining and Real Application Testing options

SQL>

2. 检查 RAC 上的 scan-ip 配置：

[grid@RAC01 ~]$ srvctl config scan

SCAN name: RACSCAN.localdomain., Network: 1/10.134.30.0/255.255.255.0/eth0

SCAN VIP name: scan1, IP: /RACSCAN.localdomain/10.134.30.50

SCAN VIP name: scan2, IP: /RACSCAN.localdomain/10.134.30.51

SCAN VIP name: scan3, IP: /RACSCAN.localdomain/10.134.30.52

[grid@RAC01 ~]$ srvctl config scan_listener

SCAN Listener LISTENER_SCAN1 exists. Port: TCP:1521

SCAN Listener LISTENER_SCAN2 exists. Port: TCP:1521

SCAN Listener LISTENER_SCAN3 exists. Port: TCP:1521

3. DNS测试：

[root@RAC01 ~]# nslookup RACSCAN.localdomain.

Server: 10.134.30.27

Address: 10.134.30.27#53

Name: RACSCAN.localdomain

Address: 10.134.30.52

Name: RACSCAN.localdomain

Address: 10.134.30.50

Name: RACSCAN.localdomain

Address: 10.134.30.51

[grid@RAC01 ~]$ nslookup 10.134.30.50

Server: 10.134.30.27

Address: 10.134.30.27#53

50.30.134.10.in-addr.arpa name = RACSCAN.localdomain.

[grid@RAC01 ~]$ nslookup 10.134.30.51

Server: 10.134.30.27

Address: 10.134.30.27#53

51.30.134.10.in-addr.arpa name = RACSCAN.localdomain.

[grid@RAC01 ~]$ nslookup 10.134.30.52

Server: 10.134.30.27

Address: 10.134.30.27#53

52.30.134.10.in-addr.arpa name = RACSCAN.localdomain.

多次运行 nslookup RACSCAN.localdomain. 可观察到 scan-ip 轮询没有任何问题，反向解析亦没有任何问题。

4. Ping三个 scan-ip 没有任何问题：

primary$ping 10.134.30.47

PING 10.134.30.47 (10.134.30.47) 56(84) bytes of data.

64 bytes from 10.134.30.47: icmp_seq=1 ttl=63 time=0.508 ms

primary$ping 10.134.30.48

PING 10.134.30.48 (10.134.30.48) 56(84) bytes of data.

64 bytes from 10.134.30.48: icmp_seq=1 ttl=63 time=0.522 ms

primary$ping 10.134.30.50

PING 10.134.30.50 (10.134.30.50) 56(84) bytes of data.

64 bytes from 10.134.30.50: icmp_seq=1 ttl=63 time=0.514 ms

5. 在任意客户端测试可否登陆：

primary$tnsping guijian

TNS Ping Utility for Linux: Version 11.2.0.1.0 – Production on 02-JAN-2014 19:12:15

Used parameter files:

/u/app/oracle/product/11g/db/network/admin/sqlnet.ora

Used TNSNAMES adapter to resolve the alias

Attempting to contact (DESCRIPTION = (ADDRESS = (PROTOCOL = TCP)(HOST = RACSCAN.localdomain.)(PORT = 1521)) (CONNECT_DATA = (SERVER = DEDICATED) (SERVICE_NAME = RACDB)))

OK (0 msec)————————-à此处透过 scan-ip 的tnsping正常

primary$

primary$sqlplus scott/Testpassword@guijian

SQL*Plus: Release 11.2.0.1.0 Production on Thu Jan 2 19:08:06 2014

ERROR:

ORA-12170: TNS:Connect timeout occurred

primary$

等待 N 久之后提示超时，但此时的 tnsping 正常，起初测试也曾提示提示 no listener。

6. 检查监听器配置：

[grid@RAC01 admin]$ cat listener.ora

LISTENER_SCAN3=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER_SCAN3)))) # line added by Agent

LISTENER_SCAN2=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER_SCAN2)))) # line added by Agent

LISTENER=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER)))) # line added by Agent

LISTENER_SCAN1=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER_SCAN1)))) # line added by Agent

LISTENER_SCAN1=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER_SCAN2)))) # line added by Agent

LISTENER_SCAN1=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER_SCAN3)))) # line added by Agent

ENABLE_GLOBAL_DYNAMIC_ENDPOINT_LISTENER_SCAN1=ON # line added by Agent

ENABLE_GLOBAL_DYNAMIC_ENDPOINT_LISTENER_SCAN2=ON # line added by Agent

ENABLE_GLOBAL_DYNAMIC_ENDPOINT_LISTENER_SCAN3=ON # line added by Agent

ENABLE_GLOBAL_DYNAMIC_ENDPOINT_LISTENER=ON # line added by Agent

监听器配置没有问题，为了方便本人讲两个节点的 listener.ora 文件内容修改为一模一样的配置。

[grid@RAC01 admin]$

LSNRCTL> status

Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER)))

STATUS of the LISTENER

————————

Alias LISTENER

Version TNSLSNR for Linux: Version 11.2.0.1.0 – Production

Start Date 02-JAN-2014 16:56:51

Uptime 0 days 1 hr. 36 min. 54 sec

Trace Level off

Security ON: Local OS Authentication

SNMP OFF

Listener Parameter File /u/app/11.2.0/grid/network/admin/listener.ora

Listener Log File /u/app/grid/diag/tnslsnr/RAC01/listener/alert/log.xml

Listening Endpoints Summary…

(DESCRIPTION=(ADDRESS=(PROTOCOL=ipc)(KEY=LISTENER)))

(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=10.134.30.27)(PORT=1521)))

(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=10.134.30.47)(PORT=1521)))

The listener supports no services

The command completed successfully

此时看到监听器 no services，其实所有的问题都是与这个no services 有关的，后续的内容中我会重点关注解决这个问题。

经过初步的检测个人感觉问题不可能出在服务器监听这一块。故此怀疑可能出在与监听相关的初始化参数上。

7. 检查数据库两个与监听相关的参数：

SQL> show parameter local_lis

NAME TYPE VALUE

———————————— ———– ——————————

local_listener string (DESCRIPTION=(ADDRESS_LIST=(AD

DRESS=(PROTOCOL=TCP)(HOST=RAC0

2-vip)(PORT=1521))))

SQL> show parameter remote_lis

NAME TYPE VALUE

———————————— ———– ——————————

remote_listener string RACSCAN.localdomain.:1521

SQL> show parameter local_lis

NAME TYPE VALUE

———————————— ———– ——————————

local_listener string (DESCRIPTION=(ADDRESS_LIST=(AD

DRESS=(PROTOCOL=TCP)(HOST=RAC0

1-vip)(PORT=1521))))

SQL> show parameter remote_lis

NAME TYPE VALUE

———————————— ———– ——————————

remote_listener string RACSCAN.localdomain.:1521

SQL>

注意上述中的 local_listener 参数的配置，发现时指向 vip 的；

8. 修改该参数指向scan-ip。

SQL> alter system set local_listener='(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=RACSCAN.localdomain.)(PORT=1521))))’;

System altered.

9. 客户端再次测试：

primary$tnsping guijian

TNS Ping Utility for Linux: Version 11.2.0.1.0 – Production on 02-JAN-2014 19:39:25

Used parameter files:

/u/app/oracle/product/11g/db/network/admin/sqlnet.ora

Used TNSNAMES adapter to resolve the alias

Attempting to contact (DESCRIPTION = (ADDRESS = (PROTOCOL = TCP)(HOST = RACSCAN.localdomain.)(PORT = 1521)) (CONNECT_DATA = (SERVER = DEDICATED) (SERVICE_NAME = RACDB)))

OK (10 msec)

primary$

primary$sqlplus scott/testpassword@guijian

SQL*Plus: Release 11.2.0.1.0 Production on Thu Jan 2 19:40:26 2014

Connected to:

Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 – 64bit Production

With the Partitioning, Real Application Clusters, Automatic Storage Management, OLAP,

Data Mining and Real Application Testing options

SCOTT@guijian>

此时可以看到顺利登陆。此外该参数修改为 scan-ip 后，在此再客户端尝试着使用 vip 登陆时开始报错：

primary$sqlplus scott/Testpassword@guijian

SQL*Plus: Release 11.2.0.1.0 Production on Thu Jan 2 19:42:54 2014