Installing Oracle Enterprise Manager 12c 12.1.0.2 on CentOS 6.3 on VMware Workstation 8

In this blog post, we are going to install Oracle Enterprise Manager 12c (12.1.0.2) on CentOS 6.3 64-bit using VMware Workstation 8.

Oracle Enterprise Manager (OEM) has been around since the 8.0 days. What started out as a fairly light-weight client-server GUI tool for managing database functions has morphed into a complex and brittle mess of poor interface design and sloppy coding.

Although 12c is a step up from the god-awful OEM 11g, the hapless Oracle DBA is still left with a muddled mess of an interface and a tool that is likely to cause more headaches than it solves.

If you are looking for Oracle 12c RAC Database install on Linux, click here

In addition the OEM system is a massive resource hog that requires an order of magnitude more horsepower, memory and disk space than the core database.

As part of this demonstration we are also going to create an Oracle 11.2.0.3 database that will be our OEM repository database. If you have a suitable repository database already installed elsewhere in your environment, you make obviously skip those steps.

I will not be installing the Grid infrastructure, as I will use the EXT4 file systems for my database datafiles, and not ASM.

Time Required: 180 minutes

Class Materials:

• An x86 64-bit computer with 8GB RAM and 60GB of free hard disk space.
• A minimum of a 1024×768 display with 256 colors.
• VMware Workstation 8 or better.
• Whatever OS VMware Workstation needs to run on. In my case I run it on Windows 7 Professional.
• Oracle Database 11.2.0.3
• Oracle Enterprise Manager 12c 12.1.0.2

You can download the software you need from the following links.

Software Download Summary:

Software Product	Download Link
VMware Workstation	VMware Workstation download
CentOS 6.3 64-bit	CentOS Linux download
Oracle 11.2.0.3 Database	Oracle 11gR2 11.2.0.3 Grid for Linux x86 64-bit Download
Oracle 12.1.0.2 Enterprise Manager Cloud Control	Oracle OEM 12c 12.1.0.2 for Linux x86 64-bit Download

 

Part I – Create the CentOS VM.

Time Required: 10 mins.

First we need to create a CentOS VM. I created a new VM using the VMware Workstation wizard. VMware Workstation 8 recognizes the CentOS 6.3 install ISO image as CentOS 64-bit and allows Easy Installer to do most of the work.

I created my VM with 6GB RAM and a single 60GB SCSI disks. The disk does not have to be fully allocated and may be split into smaller files. I have tried smaller memory settings but the performance becomes so bad it is unusable. OEM 12c is a hog.

 

Part II – Configure the Network.

Time Required: 10 mins.

In this next section we are going to add a second virtual network adapter and assign a new name and IP address to our VM.

VMware Workstation allows us to create virtual networks that are isolated from any physical network your system may reside on. Since you may or may not have control over the physical network you are connected to, this feature is important to OEM 12c where network setting predictability is important. For example I am installing this on a laptop computer and my so my laptop’s IP address changes as I travel.

Shut down your CentOS 6.3 VM. We are going to use the VMware Workstation interface and select Edit->Virtual Network Editor.

This allows us to create new virtual networks that allow our guest VMs to talk amongst themselves, to the host computer and/or to the outside world.

For this example I am going to create VMnet2.

VMnet2 will be a host-only network allowing communication to other VMs and the host, but not to the outside world. We will assign a Subnet IP of 10.10.1.0 to VMnet2, and a Subnet mask of 255.255.255.0. DHCP is enabled.

Now we have the network created in VMware, we will add a new NIC to our VM.

Right click on the VM in the VMware Workstation menu, and select Settings. This will bring up the Virtual Machine Settings panel.

Now click Add.

The Add Hardware Wizard allows us to select Network Adapter

On the Network Adapter Type menu, we will select the Custom: Specific virtual network radio button, and use the drop down to select VMnet2.

Our final VM hardware configuration should look like this:

We can now restart the VM.

Next we will use the Linux GUI interface to assign an IP address to our new NIC. Use the interface and select:

System->Preferences->Network Connections

This will bring up the Network Connections editor. You should see two adapters listed; System eth0 and a new Auto eth1. Select Auto eth1 and click Edit.

Click the IPv4 Settings tab and change the Method drop down to Manual instead of Automatic (DHCP).

Now click the Add button, then enter the IP address details. In my example I am using a static IP address on my virtual network of 10.10.1.100. My Netmask will be 255.255.255.0. Leave the Gateway at 0.0.0.0.

When you have entered the details, make sure the Require IPv4 addressing for this connection to complete box is checked, and click the Apply button.

Your final configuration should look something like this:

To assign a new name to our VM we will log in as root and edit the /etc/sysconfig/network file. In this example I am assigning the name as turandot.

[root@localhost ~]# cat /etc/sysconfig/network 
NETWORKING=yes
HOSTNAME=turandot.hadesnet

 
We need to add the network details for Turandot to the /etc/hosts file. It is important we have both the fully qualified domain name and the short name listed. If either the hostname or the domain contain the strings localhost or localdomain the install will fail. You must therefore name your host and the domain.

[root@localhost ~]# cat /etc/hosts
127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4
::1         localhost localhost.localdomain localhost6 localhost6.localdomain6
10.10.1.100 turandot.hadesnet turandot

 
Now you should reboot the VM and make sure it comes back up with the new hostname and IP address.

 

Part III – Configure the Kernel.

Time Required: 10 mins.

Next we need to configure the CentOS 6.3 Linux kernel to support Oracle 11gR2 11.2.0.3 and OEM 12c.

Edit the /etc/sysctl.conf and add following lines:

# added for Oracle 11.2.0.3 and OEM12c
kernel.shmall = 2097152
kernel.shmmax = 3066634240
kernel.shmmni = 4096
kernel.sem = 250 32000 100 128
fs.suid_dumpable = 1
fs.file-max = 6815744
fs.aio-max-nr = 1048576
net.ipv4.ip_local_port_range = 9000 65500
net.core.rmem_default = 262144
net.core.rmem_max = 4194304
net.core.wmem_default = 262144
net.core.wmem_max = 1048576

 
Remember to comment out the existing settings for kernel.shmall and kernel.shmmax. Now execute sysctl -p command to apply the new settings.

Edit the /etc/pam.d/login file and add following line:

# added for Oracle 11.2.0.3 and OEM12c
session    required     pam_limits.so

 
Edit the /etc/security/limits.conf file and add following lines:

# added for Oracle 11.2.0.3 and OEM12c
oracle    soft  nproc   2047
oracle    hard  nproc   16384
oracle    soft  nofile  1024
oracle    hard  nofile  65536
oracle    soft  stack   10240

 
Check current status of SELinux:

[root@turandot ~]# /usr/sbin/getenforce
Enforcing

 
If output is Enforcing then change mode to Permissive as follows:

[root@turandot ~]# /usr/sbin/setenforce 0

 
To make the change permanent, modify the /etc/sysconfig/selinux change value of SELINUX variable to disabled:

[root@turandot ~]# cat /etc/sysconfig/selinux

# This file controls the state of SELinux on the system.
# SELINUX= can take one of these three values:
#       enforcing - SELinux security policy is enforced.
#       permissive - SELinux prints warnings instead of enforcing.
#       disabled - SELinux is fully disabled.
SELINUX=disabled
# SELINUXTYPE= type of policy in use. Possible values are:
#       targeted - Only targeted network daemons are protected.
#       strict - Full SELinux protection.
SELINUXTYPE=targeted

 
Now we are going to disable the Linux firewall, which will otherwise interfere with the operation of OEM 12c. First let’s make sure the firewall service is stopped:

[root@turandot ~]# service iptables stop
Flushing firewall rules: [  OK  ]
Setting chains to policy ACCEPT: filter [  OK  ]
Unloading iptables modules: [  OK  ]

 
To permanently disable the firewall we will disable the service:

[root@turandot ~]# chkconfig iptables off

 
 

Part IV – Add Required RPMs.

Time Required: 10 mins.

Oracle 11gR2 and OEM 12c require extra RPMs from the install media.

The following command should load all needed RPMs for 11.2.0.3 Database and OEM 12c.

rpm -ivh compat-libstdc++-33-3.2.3-69.el6.*.rpm \
 elfutils-devel-0.152-1.el6.x86_64.rpm \
 elfutils-libelf-devel-0.152-1.el6.x86_64.rpm \
 gcc-c++-4.4.6-4.el6.x86_64.rpm \
 glibc-2.12-1.80.el6.i686.rpm \
 glibc-devel-2.12-1.80.el6.i686.rpm \
 libaio-devel-0.3.107-10.el6.x86_64.rpm \
 libaio-0.3.107-10.el6.i686.rpm \
 libgcc-4.4.6-4.el6.i686.rpm \
 libstdc++-devel-4.4.6-4.el6.x86_64.rpm \
 libtool-ltdl-2.2.6-15.5.el6.i686.rpm \
 nss-softokn-freebl-3.12.9-11.el6.i686.rpm \
 readline-6.0-4.el6.i686.rpm \
 ncurses-libs-5.7-3.20090208.el6.i686.rpm \
 libcap-2.16-5.5.el6.i686.rpm \
 libattr-2.4.44-7.el6.i686.rpm \
 compat-libcap1-1.10-1.*.rpm

 
Please note that not all the above RPMs are listed by the Oracle Installer as required. However testing has shown that they are needed to complete the install.

If you do not plan to install an Oracle database as part of this exercise, as you intend to use another host/database for the repository, then you only need the glibc-devel-2.12-1.80.el6.i686.rpm package.

With our install DVD ISO mounted to the VM, log in as root, shift to the package install directory and execute the command:

[root@turandot ~]# cd "/media/CentOS_6.3_Final/Packages/"

 
CentOS should respond with:

warning: compat-libstdc++-33-3.2.3-69.el6.i686.rpm: Header V3 RSA/SHA256 Signature, key ID c105b9de: NOKEY
Preparing...                ########################################### [100%]
   1:libstdc++-devel        ########################################### [  5%]
   2:elfutils-libelf-devel  ########################################### [ 11%]
   3:libgcc                 ########################################### [ 16%]
   4:elfutils-devel         ########################################### [ 21%]
   5:nss-softokn-freebl     ########################################### [ 26%]
   6:glibc                  ########################################### [ 32%]
   7:compat-libstdc++-33    ########################################### [ 37%]
   8:gcc-c++                ########################################### [ 42%]
   9:glibc-devel            ########################################### [ 47%]
  10:compat-libcap1         ########################################### [ 53%]
  11:libaio                 ########################################### [ 58%]
  12:ncurses-libs           ########################################### [ 63%]
  13:libattr                ########################################### [ 68%]
  14:libaio-devel           ########################################### [ 74%]
  15:libcap                 ########################################### [ 79%]
  16:readline               ########################################### [ 84%]
  17:compat-libstdc++-33    ########################################### [ 89%]
  18:libtool-ltdl           ########################################### [ 95%]
  19:compat-libcap1         ########################################### [100%]

 

If you want the pdksh RPM then you will have to download it from the web. Luckily it seems CentOS 6.3 has all the required packages to support it already, so there is no RPM dependency hell for this one.

[root@turandot ~]# rpm -ivh pdksh-5.2.14-30.x86_64.rpm
warning: pdksh-5.2.14-30.x86_64.rpm: Header V3 DSA/SHA1 Signature, key ID 73307de6: NOKEY
Preparing...                ########################################### [100%]
   1:pdksh                  ########################################### [100%]

 
 

Part V – Add the Oracle User.

Time Required: 10 mins.

Next we need to create the Oracle user account, OS groups and mount point.

The Oracle software will be installed and owned by the oracle user account. In additional several OS groups are created that allow other non oracle OS users privileged access to the database and grid resources.

Login as root and create the user oracle account and the OS groups:

[root@turandot ~]# groupadd dba
[root@turandot ~]# groupadd oinstall
[root@turandot ~]# useradd -g oinstall -G dba oracle

 
Now set the password for the oracle user. Since this is development, I suggest we stick with “oracle”.

[root@turandot ~]# echo oracle | passwd oracle --stdin
Changing password for user oracle.
passwd: all authentication tokens updated successfully.

 
Now we create a directory into which the oracle software will be installed. We need to set this new directory to be owned by oracle:

[root@turandot ~]# mkdir /u01
[root@turandot ~]# mkdir /u01/app
[root@turandot ~]# chown oracle:dba /u01/app

 
Now we can log into our new oracle account and configure the bash profile:

[root@turandot ~]#  su - oracle
[oracle@turandot ~]$  vi ~/.bash_profile

 
Add the following lines to the profile script:

TMP=/tmp; export TMP
TMPDIR=$TMP; export TMPDIR
ORACLE_HOSTNAME=`hostname`; export ORACLE_HOSTNAME
ORACLE_BASE=/u01/app/oracle; export ORACLE_BASE
ORACLE_TERM=xterm; export ORACLE_TERM
ORACLE_UNQNAME=oemctl; export ORACLE_UNQNAME
PATH=/usr/sbin:$PATH; export PATH
LD_LIBRARY_PATH=$ORACLE_HOME/lib:/lib:/usr/lib; export LD_LIBRARY_PATH
CLASSPATH=$ORACLE_HOME/JRE:$ORACLE_HOME/jlib:$ORACLE_HOME/rdbms/jlib; export CLASSPATH

if [ $USER = "oracle" ]; then
  if [ $SHELL = "/bin/ksh" ]; then
    ulimit -p 16384
    ulimit -n 65536
  else
    ulimit -u 16384 -n 65536
  fi
fi

 
 

Part VI – Install Database Software.

Time Required: 15 mins.

Next we will install the Database home.

For Oracle 11.2.0.3 this means unzipping files p10404530_112030_platform_1of7.zip and p10404530_112030_platform_2of7.zip. Note these should be unzipped into the same directory, and not separate directories as was the case in older releases.

This will create a database directory. The Database install process is fairly straightforward and now very well documented. In this example I am going to use the silent install mode using the same response file I used for my blog post Silent Install of Oracle 11.2.0.3 on RedHat 6.0.

Change into the database directory and launch the runInstaller script as follows:

[oracle@turandot database]$ ./runInstaller -silent -responseFile $HOME/db.rsp
Starting Oracle Universal Installer...

Checking Temp space: must be greater than 120 MB.   Actual 49237 MB    Passed
Checking swap space: must be greater than 150 MB.   Actual 5983 MB    Passed
Preparing to launch Oracle Universal Installer from /tmp/OraInstall2013-01-12_03-39-51PM. Please wait ...[oracle@turandot database]$ You can find the log of this install session at:
 /u01/app/oraInventory/logs/installActions2013-01-12_03-39-51PM.log

The installation of Oracle Database 11g was successful.
Please check '/u01/app/oraInventory/logs/silentInstall2013-01-12_03-39-51PM.log' for more details.

As a root user, execute the following script(s):
	1. /u01/app/oraInventory/orainstRoot.sh
	2. /u01/app/oracle/product/11.2.0/dbhome_1/root.sh


Successfully Setup Software.

 
At the conclusion of the install we need to log in as root and execute the two scripts shown.

My complete db.rsp file can be found here: db.rsp. WordPress does not allow straight text files to be uploaded, so the extension is DOC, although it is a raw ASCII text file.

 

Part VII – Create Listener and Database.

Time Required: 20 mins.

The OEM 12c system needs an Oracle database to use as a repository. Since I want this VM to be self-contained for my own educational purposes, I am going to add a database to serve that purpose.

Before we create a database we need to configure the listener. To simplify matters I find it a good idea to add a dummy database entry to the /etc/oratab file. Then I can use the oraenv script to source the environment, and then use the Oracle binaries to complete the setup.

[oracle@turandot database]$ . oraenv
ORACLE_SID = [oracle] ? DUMMY
The Oracle base has been set to /u01/app/oracle

[oracle@turandot database]$ netca -silent -responsefile $ORACLE_HOME/assistants/netca/netca.rsp

Parsing command line arguments:
    Parameter "silent" = true
    Parameter "responsefile" = /u01/app/oracle/product/11.2.0/dbhome_1/assistants/netca/netca.rsp
Done parsing command line arguments.
Oracle Net Services Configuration:
Profile configuration complete.
Oracle Net Listener Startup:
    Running Listener Control: 
      /u01/app/oracle/product/11.2.0/dbhome_1/bin/lsnrctl start LISTENER
    Listener Control complete.
    Listener started successfully.
Listener configuration complete.
Oracle Net Services configuration successful. The exit code is 0

 
Now we can create a database from the DBCA template General Purpose. In this example we set the SYSTEM and SYSDBA passwords to oracle, use the file system for storage and name the database oemctl.

We are also setting the following parameters to comply with OEM 12c requirements:

• processes=300
• session_cached_cursors=300
• shared_pool_size=600M
• redolog size 300M

The new database will register with the local listener LISTENER, which was created in the previous step.

dbca -silent \
  -createDatabase \
  -characterSet WE8ISO8859P15 \
  -continueOnNonFatalErrors false \
  -datafileDestination '/u01/app/oracle/oradata' \
  -disableSecurityConfiguration ALL \
  -emConfiguration NONE \
  -gdbName oemctl.local \
  -initParams processes=300,session_cached_cursors=300,shared_pool_size=600M \
  -listeners LISTENER \
  -memoryPercentage 40 \
  -recoveryAreaDestination '/u01/app/oracle/flash_recovery_area' \
  -redoLogFileSize 300 \
  -sid oemctl \
  -SysPassword oracle \
  -SystemPassword oracle \
  -storageType FS \
  -sampleSchema false \
  -templateName General_Purpose.dbc

Copying database files
1% complete
3% complete
11% complete
18% complete
26% complete
37% complete
Creating and starting Oracle instance
40% complete
45% complete
50% complete
55% complete
56% complete
60% complete
62% complete
Completing Database Creation
66% complete
70% complete
73% complete
85% complete
96% complete
100% complete
Look at the log file "/u01/app/oracle/cfgtoollogs/dbca/oemctl/oemctl.log" for further details.

 
Set the database to auto-start by editing the /etc/oratab file. We will be using this feature later on.

[oracle@turandot database]$ cat /etc/oratab
#
oemctl:/u01/app/oracle/product/11.2.0/dbhome_1:Y		# line added by
#

 
Even though we set the emConfiguration to NONE, DBCA still creates an OEM repository schema, which will cause the OEM installer to have a temper tantrum.

Before we go any further we need to fully deconfigure and drop the OEM/DBC components with the following command:

emca -deconfig dbcontrol db -repos drop -silent \
 -SID oemctl \
 -PORT 1521 \
 -LISTENER_OH /u01/app/oracle/product/11.2.0/dbhome_1 \
 -SYS_PWD oracle \
 -DBSNMP_PWD dbsnmp \
 -SYSMAN_PWD oem_temp

 
Executing the command should produce the following output:

STARTED EMCA at Jan 12, 2013 4:25:43 PM
EM Configuration Assistant, Version 11.2.0.3.0 Production
Copyright (c) 2003, 2011, Oracle.  All rights reserved.


----------------------------------------------------------------------
WARNING : While repository is dropped the database will be put in quiesce mode.
----------------------------------------------------------------------
Jan 12, 2013 4:25:43 PM oracle.sysman.emcp.EMConfig perform
INFO: This operation is being logged at /u01/app/oracle/cfgtoollogs/emca/oemctl/emca_2013_01_12_16_25_42.log.
Jan 12, 2013 4:25:44 PM oracle.sysman.emcp.util.GeneralUtil initSQLEngineLoacly
WARNING: ORA-28000: the account is locked

Jan 12, 2013 4:25:44 PM oracle.sysman.emcp.ParamsManager checkListenerStatusForDBControl
WARNING: Error initializing SQL connection. SQL operations cannot be performed
Jan 12, 2013 4:25:44 PM oracle.sysman.emcp.EMDBPreConfig performDeconfiguration
WARNING: EM is not configured for this database. No EM-specific actions can be performed. Some of the possible reasons may be: 
 1) EM is configured with different hostname then physical host. Set environment variable ORACLE_HOSTNAME= and re-run EMCA script 
 2) ORACLE_HOSTNAME is set. Unset it and re-run EMCA script
Jan 12, 2013 4:25:44 PM oracle.sysman.emcp.EMReposConfig invoke
INFO: Dropping the EM repository (this may take a while) ...
Jan 12, 2013 4:26:46 PM oracle.sysman.emcp.EMReposConfig invoke
INFO: Repository successfully dropped
Enterprise Manager configuration completed successfully
FINISHED EMCA at Jan 12, 2013 4:26:46 PM

 
It is good practice to check the database to ensure that both the SYSMAN and MGMT_VIEW users as well as the MGMT_USER role have been dropped.

 

Part VIII – Installing OEM 12c.

Time Required: 60 mins.

The OEM12c is available for download from download.oracle.com as three separate zip files. You need all three:

• em12cr2_linux64_disk1.zip
• em12cr2_linux64_disk2.zip
• em12cr2_linux64_disk3.zip

All zip files must be unzipped into the same directory.

If like me you are using VMware Workstation on a laptop with limited disk space, then I recommend you use a plug-in USB stick for your install media, that way you don’t need to provision additional disk space to stage the install files.

Note! The installer is extremely brittle. It cannot execute from a directory path with spaces in it. If your plug-in USB stick uses spaces in the directory name, you will need to change to remove spaces that before starting.

Note! – do NOT use the runInstaller in the base directory. That version does not work! You will find a second runInstaller in the install sub-directory.

Launch the installer and follow the prompts:

Step 1 of 9

Uncheck the I wish to recieve security updates via My Oracle Support and click Next. Dismiss the warning dialog.

Step 2 of 9

Check Skip for Software Updates. Click Next.

Step 3 of 9

All pre-requisite checks should pass. If anything was missed it must be corrected before proceeding.

Step 4 of 9

Ensure the Create a new Enterprise Manager System option is checked, and select Advanced. Click Next.

Step 5 of 13

Set the install directories as follows:

Component	Location
Middleware Home Location	/u01/app/oracle/Middleware
Agent Base Directory	/u01/app/oracle/agent

Step 6 of 13

Accept the default plug-in list.

Step 7 of 13

Enter passwords for the weblogic and nodemanager users. Write down your selections as you will need them later.

The password must be eight characters and include one numeric.

Accept the default install location for the OMS Instance Base Location.

Step 8 of 13

Set the Database Connection Details as follows:

Item	Setting
Database Host Name	turandot
Port	1521
Service/SID	oemctl
SYS Password	oracle

The installer will complain about the target database and offer to disable the stats gathering job to correct the pre-req. Click OK.

Step 9 of 13

Enter your password selection for SYSMAN and Registration accounts.

Accept the default tablespace settings for Management, Configuration Data and Diagnostics Data.

Step 10 of 13

Accept the default port assignments.

Step 11 of 13

Review the configuration data to ensure everything is accurate.

Click Install.

The install process is very slow. Relax, grab a drink and put on some music.

Step 12 of 13

When prompted, execute the specified script using the root account:

[root@turandot oms]# ./allroot.sh

Starting to execute allroot.sh ......... 

Starting to execute /u01/app/oracle/Middleware/oms/root.sh ......
Running Oracle 11g root.sh script...

The following environment variables are set as:
    ORACLE_OWNER= oracle
    ORACLE_HOME=  /u01/app/oracle/Middleware/oms

Enter the full pathname of the local bin directory: [/usr/local/bin]: 
The file "dbhome" already exists in /usr/local/bin.  Overwrite it? (y/n) 
[n]: y
   Copying dbhome to /usr/local/bin ...
The file "oraenv" already exists in /usr/local/bin.  Overwrite it? (y/n) 
[n]: y
   Copying oraenv to /usr/local/bin ...
The file "coraenv" already exists in /usr/local/bin.  Overwrite it? (y/n) 
[n]: y
   Copying coraenv to /usr/local/bin ...

Entries will be added to the /etc/oratab file as needed by
Database Configuration Assistant when a database is created
Finished running generic part of root.sh script.
Now product-specific root actions will be performed.
/etc exist

Creating /etc/oragchomelist file...
/u01/app/oracle/Middleware/oms
Finished execution of  /u01/app/oracle/Middleware/oms/root.sh ......


Starting to execute /u01/app/oracle/agent/core/12.1.0.2.0/root.sh ......
Finished product-specific root actions.
/etc exist
Finished execution of  /u01/app/oracle/agent/core/12.1.0.2.0/root.sh ......

 
Click the OK button to complete the install.

Step 13 of 13

The final screen shows a summary of URLs to use to connect to OEM Cloud Control. Make a note of these for later use.

Click the Close button to exit the installer.

 

Part IX – Setting Database and OEM to Autostart.

Time Required: 10 mins.

With the introduction of Oracle Restart which uses the Grid infrastructure, Oracle no longer installs the dbstart script to auto-start the database. Since we did not install the Grid Infrastructure, we need our own auto-start script, to start up the listener and database.

Oracle does install the script /etc/init.d/gcstartup which starts the OMS server and Agent, but the OMS start will fail if the repository database is not up.

Furthermore, the /etc/init.d/gcstartup script is as much use as a chocolate tea-pot. Even if the database is up, it usually starts the OMS server and leaves the WebTier down, rendering your OEM system useless.

The tag line of this blog is When you just want this Oracle @&$! to work!. Therefore we are going to create our own startup script to handle the database, listener, OEM and the agent.

The following script should be created by the root user as /etc/init.d/dbstartup.

[root@turandot ~]# cat /etc/init.d/dbstartup
#!/bin/bash

ORA_OWNER=oracle
export ORACLE_HOME=/u01/app/oracle/product/11.2.0/dbhome_1
export OMS_HOME=/u01/app/oracle/Middleware/oms
export AGENT_HOME=/u01/app/oracle/agent/core/12.1.0.2.0

case "$1" in
    'start')
        # Start everything
        su - $ORA_OWNER -c "$ORACLE_HOME/bin/dbstart $ORACLE_HOME"
        su - $ORA_OWNER -c "$OMS_HOME/bin/emctl start oms"
        su - $ORA_OWNER -c "$AGENT_HOME/bin/emctl start agent"
        ;;

    'stop')
        # Stop everything
        su - $ORA_OWNER -c "$OMS_HOME/bin/emctl stop oms -all"
        su - $ORA_OWNER -c "$AGENT_HOME/bin/emctl stop agent"
        su - $ORA_OWNER -c "$ORACLE_HOME/bin/dbshut $ORACLE_HOME"
        ;;

esac

 
Now set the correct permission on the file:

[root@turandot ~]# chmod 750 /etc/init.d/dbstartup

 
We now need to create symbolic links from the run-level directories as follows:

ln -s /etc/rc.d/init.d/dbstartup /etc/rc.d/rc2.d/S99dbstartup
ln -s /etc/rc.d/init.d/dbstartup /etc/rc.d/rc3.d/S99dbstartup
ln -s /etc/rc.d/init.d/dbstartup /etc/rc.d/rc4.d/S99dbstartup
ln -s /etc/rc.d/init.d/dbstartup /etc/rc.d/rc5.d/S99dbstartup

ln -s /etc/rc.d/init.d/dbstartup /etc/rc.d/rc0.d/K99dbstartup # Halting
ln -s /etc/rc.d/init.d/dbstartup /etc/rc.d/rc6.d/K99dbstartup # Rebooting

 
And we need to delete or cripple the /etc/init.d/gcstartup script.

This will ensure the entire Oracle stack starts and stops with the server.

 

Part X – Connecting to the OEM Interface.

Time Required: 20 mins.

You can connect to your new OMS with a flash enabled web browser within the VM or from your host computer. Point a browser at https://servername:7802/em.

The user name is SYSMAN and the password is whatever you specified during the install.

Since this is a test system, we are going to add our oemctl database as a monitoring target for OEM. In order to do this we need to first unlock the dbsnmp account in the oemctl database.

SQL> alter user dbsnmp account unlock;

User altered.

SQL> alter user dbsnmp identified by dbsnmp;

User altered.

 
Now point a browser at the OEM log on page:

Click on Targets to bring up the current known list of targets. You will note our oemctl database is not listed.

Click on Setup->Add Target->Add Targets Manually to launch the dialog to add a new target database.

In the Add Targets Manually page, select Add Non-Host Targets Using Guided Process (Also Adds Related Targets) and then select Oracle Database, Listener and Automatic Storage Management from the drop-down list.

In theh Add Database Instance Target: Specify Host page, enter the hostname.

The next page tries to discover targets on the specified host. Here we see it has found our oemctl database, and asks for a monitoring password.

Enter the password for the dbsnmp account in the Monitor Password box, and click Test Connection. The test should be successful.

The next page allows us to record some details about the monitor target, since this is a sandbox, it is not important.

The last screen is a summary screen.

Back on the All Targets screen we can see the newly added target and the status is shown.