Goldengate Monitor安装时TS节选剧情

最近想研究下Goldengate Monitor。所以就在linux上搭了一个ogg的环境，准备部署Goldengate Monitor 11.2。
手头有的就是一个11.1的官方手册，网上的资源也是甚少。这里想分享下我在配置Goldengate Monitor 11.2的trouble shooting的刺激过程。
直接开门见错误吧：

1、start jagent后立即stopped
[code]GGSCI (localhost.localdomain) 5> info all

Program Status Group Lag at Chkpt Time Since Chkpt

MANAGER RUNNING
JAGENT STOPPED
EXTRACT STOPPED AWEN_EXT 00:00:00 28:21:32
EXTRACT STOPPED AWEN_PMP 00:00:00 28:21:31
REPLICAT STOPPED AWEN_REP 00:00:00 28:21:30


GGSCI (localhost.localdomain) 6> view report JAGENT


2013-08-23 21:55:27 INFO OGG-00978 GGCMD JAGENT is running.

2013-08-23 21:55:27 INFO OGG-01927 Child process started, process ID 3094,
command line 'java -jar -Xms64m -Xmx512m dirjar/jagent.jar'.

2013-08-23 21:55:28 INFO OGG-01928 Child process terminated with exit code 1.

2013-08-23 21:55:28 INFO OGG-00979 GGCMD JAGENT is down (gracefully).[/code]
第一时间在官方文档和MOS上搜寻相关错误，未果。

2、后来就goggle了一把，在Roger的blog上说是java版本太低。
在官方的手册上有看到【Install a release of Java 1.6 Java Development Kit (JDK) on each system where Oracle GoldenGate is installed.】
所以在安装的时候安装了1.6的jdk，且java -version显示为1.6。

3、后来发现Roger装的是jre，但是官方手册上写着【 This must be the JDK, not a Java Runtime Environment (JRE)】，但是还是想试一试，就卸载了jdk，装了1.7的ire，错误依旧。

4、再后来干脆安装了1.7的jdk。
[oracle@localhost ~]$ java -version
java version "1.7.0_25"
Java(TM) SE Runtime Environment (build 1.7.0_25-b15)
Java HotSpot(TM) Client VM (build 23.25-b01, mixed mode, sharing)

5、且在创建密钥的时候报如下错误
[oracle@localhost ogghome]$ ./pw_agent_util.sh -create
Error: Could not find or load main class com.goldengate.monitor.jagent.security.PWAgentUtil

6、想到还是和JAVA有关，然后就建了一个test.java测试下java。【本人不懂JAVA】
[code][oracle@localhost ogghome]$ cat test.java
public class Test{
public static void main(String[] args){
System.out.println("hello java!");
}
}
[oracle@localhost ogghome]$ java test.java
Error: Could not find or load main class test.java[/code]
和之前执行那个脚本报同样的错误。网上基本上都是说环境变量的问题，我确认了几次，都没问题。
后来用看到用javac来编译
[oracle@localhost ogghome]$ javac test.java
没有报错，且生成了Test.class文件
[oracle@localhost ogghome]$ java Test

hello java!
那，到这里确认java是没问题的。
7、后来想到修改脚本pw_agent_util.sh，看到最后有一行
$JAVA_HOME/bin/java com.goldengate.monitor.jagent.security.PWAgentUtil $*
就把java改成了javac，再执行报权限不够的错误
[code][oracle@localhost ogghome]$ ./pw_agent_util.sh -create
error: error reading /u01/ogg/ogghome/dirjar/jagent.jar; /u01/ogg/ogghome/dirjar/jagent.jar (Permission denied)
error: error reading /u01/ogg/ogghome/dirjar/monitor-common.jar; /u01/ogg/ogghome/dirjar/monitor-common.jar (Permission denied)
error: error reading /u0[root@localhost ~]# cd /u01/ogg/ogghome/dirjar/1/ogg/ogghome/dirjar/commons-codec-1.3.jar; /u01/ogg/ogghome/dirjar/commons-codec-1.3.jar (Permission denied)
······省略部分
[oracle@localhost ogghome]$ ls -l /u01/ogg/ogghome/dirjar/*
-rw-r----- 1 94110 42424 46725 Apr 23 2012 /u01/ogg/ogghome/dirjar/commons-codec-1.3.jar
-rw-r----- 1 94110 42424 38015 Apr 23 2012 /u01/ogg/ogghome/dirjar/commons-logging-1.0.4.jar
······省略部分[/code]
8、发现oracle的确没有权限，之后就用root用户改了所有者，所属组
[root@localhost ~]# cd /u01/ogg/ogghome/dirjar/
[root@localhost dirjar]# chown oracle:dba *
再切回到oracle执行脚本
[oracle@localhost ogghome]$ ./pw_agent_util.sh -create
error: Class names, 'com.goldengate.monitor.jagent.security.PWAgentUtil,create', are only accepted if annotation processing is explicitly requested
1 error

9、根据错误提示，想到之前把java改为了javac ,所以改了回来，再执行
[code][oracle@localhost ogghome]$ ./pw_agent_util.sh -create
Please create a password for Java Agent:
Please confirm password for Java Agent:
Please enter Monitor Server JMX password:
Please confirm Monitor Server JMX password:
Aug 23, 2013 10:17:19 PM oracle.security.jps.internal.config.xml.XmlConfigurationFactory initDefaultConfiguration
SEVERE: java.io.FileNotFoundException: /u01/ogg/ogghome/cfg/jps-config-jse.xml (Permission denied)
0 [main] ERROR com.goldengate.wallet.WalletHelper - oracle.security.jps.config.JpsConfigurationException: /u01/ogg/ogghome/cfg/jps-config-jse.xml (Permission denied)[/code]
10、执行后，开始提示输入密码了，看到了曙光。不过后面还是报了权限不够的错误，之后干脆把ogg的目录都改了下所有者和所属组
[code][root@localhost ~]# chown -R oracle:dba /u01/ogg
[oracle@localhost ogghome]$ ./pw_agent_util.sh -create
Please create a password for Java Agent:
Please confirm password for Java Agent:
Please enter Monitor Server JMX password:
Please confirm Monitor Server JMX password:
Wallet is created successfully.[/code]
再切到oracle用户执行，bingo。
11、以最快的速度连到ogg，启动jagent。还是报错了。
[code]2013-08-23 22:20:47 INFO OGG-00978 GGCMD JAGENT is running.

2013-08-23 22:20:47 INFO OGG-01927 Child process started, process ID 3421,
command line 'java -jar -Xms64m -Xmx512m dirjar/jagent.jar'.

2013-08-23 22:20:50 INFO OGG-01928 Child process terminated successfully.

2013-08-23 22:20:50 INFO OGG-00979 GGCMD JAGENT is down (gracefully).[/code]
12、不过这次的报错不一样。之后就想单独执行java -jar -Xms64m -Xmx512m dirjar/jagent.jar这条语句看看
[code][oracle@localhost ogghome]$ java -jar -Xms64m -Xmx512m dirjar/jagent.jar
······省略
726 [main] ERROR com.goldengate.monitor.jagent.config.impl.AgentInfoImpl - Property jagent.username is not defined.
1199 [main] WARN com.goldengate.monitor.jagent.config.impl.AgentInfoImpl - Property maximum.message.retrieval is not defined. Default value 500 will be used. Define this property with a value if the default is not what you want.
1202 [main] ERROR com.goldengate.monitor.jagent.JAgentWSMain - Error Initializing JAgent. JAgent will not be initialized and about to return. Exception: org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'agentBootstrapper': Injection of resource dependencies failed; nested exception is org.springframework.beans.factory.BeanCreationException: Error creating bean with name 'agentInfo': Invocation of init method failed; nested exception is java.lang.Exception: One ore more Java Agent properties in Config.properties file are not defined properly. Check ggserr.log file for more detail. Correct the error in the Config.properties file and restart the Manager.
1202 [main] INFO com.goldengate.monitor.jagent.JAgentWSMain - JAgentWSMain is about to terminate its operation
1203 [Shutdown Thread.] INFO com.goldengate.monitor.jagent.JAgentWSMain - JAgent is about to shutdown.
1204 [Shutdown Thread.] INFO com.goldengate.monitor.jagent.JAgentWSMain - JAgent shutdown process is completed successfully.[/code]
13、应该是有参数没有设置，后来就检查Config.properties文件，在未设置的参数里面发现jagent.username比较可疑，所以就加上了这个参数。
[code][oracle@localhost cfg]$ vim Config.properties
jagent.host =127.0.0.1
monitor.host =192.168.3.142
monitor.jmx.username =jmxuser
agent.type.enabled=OGGMON
jagent.username=oggmonitor[/code]
14、再运行，发现可以运行了，但是有错误
[code]······省略
JAgent is unable to launch at JMX port: 5555.JAgent JMX port is not available.
Please change 'jagent.jmx.port' property in Config.properties file to an available port number.
2212 [main] INFO com.goldengate.monitor.jagent.JAgentWSMain - JAgentWSMain is about to terminate its operation
3095 [ManagerConnectionKeeper] INFO com.goldengate.monitor.jagent.comm.ws.ManagerWSApi - Created WSAPI[/code]
15、后来重启了monitor，还是一样
[code][oracle@localhost bin]$ ./monitor.sh stop
Using CATALINA_BASE: /u01/ogg/oggmonitor/tomcat
Using CATALINA_HOME: /u01/ogg/oggmonitor/tomcat
Using CATALINA_TMPDIR: /u01/ogg/oggmonitor/tomcat/temp
Using JRE_HOME: /u01/ogg/oggmonitor/jre
[oracle@localhost bin]$ ./monitor.sh start
Using CATALINA_BASE: /u01/ogg/oggmonitor/tomcat
Using CATALINA_HOME: /u01/ogg/oggmonitor/tomcat
Using CATALINA_TMPDIR: /u01/ogg/oggmonitor/tomcat/temp
Using JRE_HOME: /u01/ogg/oggmonitor/jre[/code]
16、登录monitor的web界面 http://192.168.3.142:5555/monitor/
果然，页面报错，tomcat找不到monitor资源，后来检查tomcatlog发现如下报错
[code]Caused by: javax.persistence.PersistenceException: Exception [EclipseLink-4002] (Eclipse Persistence Services - 2.1.2.v20101206-r8635): org.eclipse.per
sistence.exceptions.DatabaseException
Internal Exception: org.apache.commons.dbcp.SQLNestedException: Cannot create PoolableConnectionFactory (Io exception: The Network Adapter could not es
tablish the connection)
Error Code: 0[/code]
17、想到monitor是要连接oracle的资料库的，可能与监听有关，导致连接不到数据库。
[code][oracle@localhost ~]$ lsnrctl status
监听未打开
[oracle@localhost ~]$ lsnrctl start
[oracle@localhost ~]$ sqlplus / as sysdba
SQL> alter system register; [/code]
之后再登录web界面，果然好了。进ggsci打开jagent，现在终于一直running了。
[code]GGSCI (localhost.localdomain) 36> info all

Program Status Group Lag at Chkpt Time Since Chkpt

MANAGER RUNNING
JAGENT RUNNING
EXTRACT STOPPED AWEN_EXT 00:00:00 00:02:57
EXTRACT STOPPED AWEN_PMP 00:00:00 00:02:56
REPLICAT STOPPED AWEN_REP 00:00:00 00:02:54


GGSCI (localhost.localdomain) 37> view report JAGENT


***********************************************************************
Oracle GoldenGate Command Tool for Oracle
Version 11.2.1.0.1 OGGCORE_11.2.1.0.1_PLATFORMS_120423.0230
Linux, x86, 32bit (optimized) on Apr 23 2012 04:36:58

Copyright (C) 1995, 2012, Oracle and/or its affiliates. All rights reserved.


Starting at 2013-08-23 23:00:48
***********************************************************************

Operating System Version:
Linux
Version #1 SMP Tue Aug 18 15:51:54 EDT 2009, Release 2.6.18-164.el5
Node: localhost.localdomain
Machine: i686
soft limit hard limit
Address Space Size : unlimited unlimited
Heap Size : unlimited unlimited
File Size : unlimited unlimited
CPU Time : unlimited unlimited

Process id: 4138

Parameters...

COMMAND java -jar -Xms64m -Xmx512m dirjar/jagent.jar
Program is 'java', arguments are '-jar -Xms64m -Xmx512m dirjar/jagent.jar'


***********************************************************************
** Run Time Messages **
***********************************************************************


2013-08-23 23:00:48 INFO OGG-00978 GGCMD JAGENT is running.

2013-08-23 23:00:48 INFO OGG-01927 Child process started, process ID 4142, command line 'java -jar -Xms64m -Xmx512m dirjar/jagent.jar'.


GGSCI (localhost.localdomain) 38>[/code]
18、看似已经OK了，但是问题又来了。
登录到web界面，什么都没有显示，到ogghome目录下看jagent.log，里面记录的还是一个小时之前的信息，重启了monitor几次，web里面还是什么都没有。

19、后来在ggsci里面重启jagent进程，再看jagent.log的时候发现问题了。
[code]2013-08-24 00:32:41 [main] WARN AgentInfoImpl - Property jagent.jmx.port is not defined. Default value 5555 will be used. Define this property with a value if the default is not what you want.
2013-08-24 00:32:43 [main] INFO ManagerFacadeImpl - Metadata initialized
2013-08-24 00:32:43 [main] INFO JAgentDigestMD5JmxFactory - Starting JMX connector server on port 5555
2013-08-24 00:32:43 [main] ERROR MBeansContainerImpl -
JAgent is unable to launch at JMX port: 5555.JAgent JMX port is not available.
Please change 'jagent.jmx.port' property in Config.properties file to an available port number.
java.net.BindException: Address already in use [/code]
20、我最开始也没注意这个参数，因为这个端口已经给了tomcat做了http的端口。所以这里要重新配置一个新的端口，
在/ogghome/cfg/Config.properties文件中加入如下内容：
jagent.jmx.port=5588
重启monitor
[code][oracle@localhost bin]$ ./monitor.sh stop
Using CATALINA_BASE: /u01/ogg/oggmonitor/tomcat
Using CATALINA_HOME: /u01/ogg/oggmonitor/tomcat
Using CATALINA_TMPDIR: /u01/ogg/oggmonitor/tomcat/temp
Using JRE_HOME: /u01/ogg/oggmonitor/jre
[oracle@localhost bin]$ ./monitor.sh start
Using CATALINA_BASE: /u01/ogg/oggmonitor/tomcat
Using CATALINA_HOME: /u01/ogg/oggmonitor/tomcat
Using CATALINA_TMPDIR: /u01/ogg/oggmonitor/tomcat/temp
Using JRE_HOME: /u01/ogg/oggmonitor/jre[/code]
重启jagent
[code]GGSCI (localhost.localdomain) 55> stop JAGENT
JAgent process is required by other GGS processes.
Are you sure you want to stop it (y/n)? y

Sending STOP request to GGCMD JAGENT ...
Request processed.
JAgent stopped.


GGSCI (localhost.localdomain) 56> start JAGENT

GGCMD JAGENT started


GGSCI (localhost.localdomain) 57> info all

Program Status Group Lag at Chkpt Time Since Chkpt

MANAGER RUNNING
JAGENT RUNNING
EXTRACT RUNNING AWEN_EXT 00:00:00 00:00:00
EXTRACT RUNNING AWEN_PMP 00:00:00 00:00:07
REPLICAT RUNNING AWEN_REP 00:00:00 00:00:04[/code]
到这里整个TS过程结束，web界面已经可以正常显示整个拓扑结构，感觉这东西挺棒的，特别是针对多套ogg的用户，管理监控都特别方便，后面我会给出详细的安装文档。
附件为界面的截图：[img]http://www.orasql.com/blog//uploadfiles/ogg_mo_38185.jpg[/img]