Switch branches/tags
Nothing to show
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
244 lines (145 sloc) 12.2 KB

ATG CRS Quickstart Guide

Product versions used in this guide:

  • Oracle Linux Server release 6.5 (Operating System)
  • Oracle Database Enterprise Edition
  • Oracle ATG Web Commerce 11.1
  • JDK 1.7
  • ojdbc7.jar driver
  • Jboss EAP 6.1


This document describes a quick and easy way to install and play with ATG CRS. By following this guide, you'll be able to focus on learning about ATG CRS, without debugging common gotchas. This project contains a vagrant definition for a database vm. If you follow the instructions in this guide you'll have a "db" vm running Oracle Enterprise Edition, with a system username/password combo of system/oracle. This is optional. If you have another db you'd like to use, that's fine. Just make sure it's Oracle 11.2 or later. The benefit to running your oracle db in a VM is that you've got complete control over it, and you can set up and tear it down at will without worrying that you'll accidentally delete other peoples' data. Plus you know it's guarateed to work because you created it with a fresh install from scratch.

You need to run these vms on a pretty robust machine - the ATG vm requires 8 gigs of ram to run, and the DB needs 2 gigs.

You will need to refer to The ATG CRS Installation and Configuration Guide frequently after initial environment setup. A lot of what this doc does is help you through the process of meeting the installation requirements.


Throughout this document, the top-level directory that you checked out from git will be referred to as ${ATG-CRS}

Install Required Virtual Machine Software

Install the latest versions of VirtualBox and Vagrant

Download Required Database Software (optional)

The first step is to download the required installers. In order to download Oracle database software you need an Oracle Support account.

Oracle Enterprise Edition

  • Go to Oracle Support
  • Click the "patches and updates" tab
  • On the left of the page look for "patching quick links". If it's not expanded, expand it.
  • Within that tab, under "Oracle Server and Tools", click "Latest Patchsets"
  • This should bring up a popup window. Mouse over Product->Oracle Database->Linux x86-64 and click on
  • At the bottom of that page, click the link "13390677" within the table, which is the patch number
  • Only download parts 1 and 2.

Even though it says it's a patchset, it's actually a full product installer.

IMPORTANT: Put the zip files parts 1 and 2, in the ${ATG-CRS}/software directory at the top level of this project (it's the directory that has a readme.txt file telling you how to use the directory).

Oracle SQL Developer

You will also need a way to connect to the database. I recommend Oracle SQL Developer.

Build the database vm (optional)

In the root directory of this project (the one that contains the Vagrantfile), type:

$ vagrant up db

This will set in motion an amazing series of events, and will take some time (~10-15 min, depending on download and machine speeds):

  • download an empty centos machine
  • switch it to Oracle Linux (an officially supported platform for Oracle 11g and ATG 11.1)
  • install all prerequisites for Oracle
  • install and configure the oracle db software
  • create an empty db name orcl

To get a shell on the db vm, type

$ vagrant ssh db

You'll be logged in as the user "vagrant". This user has sudo privileges (meaning you can run somecommand as root by typing $ sudo somecommand). To su to root (get a root shell), type su -. The root password is "vagrant". If you want to su to the oracle user, the easiest thing to do is to su to root and then type su - oracle. The "oracle" user is the user that's running oracle and owns all the oracle directories. The project directory (the directory from which you ran vagrant up db) will be mounted at /vagrant. You can copy files back and forth between your host machine and the VM using that directory.

Key Information:

  • The db vm has the private IP This is defined at the top of the Vagrantfile.
  • The system username password combo is system/oracle
  • The SID (database name) is orcl
  • It's running on the default port 1521
  • You can control the oracle server with a service: "sudo service dbora stop|start"

Create DB schemas (Do this even if you're using your own DB, and not the one created in the steps above)

Before you can set up ATG to connect to your database, you must create four new schemas. Just make the password the same as the user name:

  • crs_core
  • crs_pub
  • crs_cata
  • crs_catb

Download required ATG server software

ATG 11.1

  • Go to Oracle Edelivery
  • Accept the restrictions
  • On the search page Select the following options:
    • Product Pack -> ATG Web Commerce
    • Platform -> Linux x86-64
  • Click Go
  • Click the top search result "Oracle Commerce (11.1.0), Linux"
  • Download the following parts:
    • Oracle Commerce Platform 11.1 for UNIX
    • Oracle Commerce Reference Store 11.1 for UNIX
    • Oracle Commerce MDEX Engine 6.5.1 for Linux
    • Oracle Commerce Content Acquisition System 11.1 for Linux
    • Oracle Commerce Experience Manager Tools and Frameworks 11.1 for Linux
    • Oracle Commerce Guided Search Platform Services 11.1 for Linux

JDK 1.7

JBoss EAP 6.1

OJDBC Driver

All oracle drivers are backwards compatible with the officially supported database versions at the time of the driver's release. You can use ojdbc7 to connect to either 12c or 11g databases.

IMPORTANT: Move everything you downloaded to the ${ATG-CRS}/software directory at the top level of this project.

Create the "atg" vm

$ vagrant up atg

When it's done you'll have a vm created that is all ready to install and run ATG CRS. It will have installed jdk7 at /usr/java/jdk1.7.0_67 and jboss at /home/vagrant/jboss/. You'll also have the required environment variables set in the .bash_profile of the "vagrant" user.

To get a shell on the atg vm, type

$ vagrant ssh atg

Key Information:

  • The atg vm has the private IP This is defined at the top of the Vagrantfile.
  • java is installed in /usr/java/jdk1.7.0_67
  • jboss is installed at /home/vagrant/jboss
  • Your project directory is mounted at /vagrant. You'll find the installers you downloaded at /vagrant/software from within the atg vm

Install Endeca (Oracle Commerce Guided Search Platform)


ssh into the atg vagrant vm. From the directory that contains this project's Vagrantfile type:

$ vagrant ssh atg

change to the directory where your installers are:

[vagrant@localhost ~]$ cd /vagrant/software

Follow the Getting Started Guide. Accept all the default values.


[vagrant@localhost software]$ /vagrant/software/ --target /usr/local

Accept all defaults and then perform the post-installation procedure

[vagrant@localhost software]$ source /usr/local/endeca/MDEX/6.5.1/mdex_setup_sh.ini

Platform services

[vagrant@localhost software]$ /vagrant/software/OCplatformservices11.1.0-Linux64.bin --target /usr/local/

[vagrant@localhost software]$ source /usr/local/endeca/PlatformServices/workspace/setup/installer_sh.ini

Tools and frameworks

You have to unzip the file into your /vagrant/software directory. This will create the folder "/vagrant/software/cd". You can't run the basic installer because it wants a display to be able to show you all their fancy dialogs and whatnot. Just run the installer silently using the provided response file. This is the same thing as accepting all the defaults.

[vagrant@localhost software]$ export ENDECA_TOOLS_ROOT=/usr/local/endeca/ToolsAndFrameworks/11.1.0

[vagrant@localhost software]$ export ENDECA_TOOLS_CONF=/usr/local/endeca/ToolsAndFrameworks/11.1.0/server/workspace

[vagrant@localhost software]$ /vagrant/software/cd/Disk1/install/ /vagrant/scripts/atg/endeca_toolsandframeworks_silent_response.rsp ToolsAndFrameworks /usr/local/endeca/ToolsAndFrameworks admin

[vagrant@localhost software]$ sudo /home/vagrant/oraInventory/


[vagrant@localhost software]$ /vagrant/software/ --target /usr/local

Start the services:


[vagrant@localhost software]$ /usr/local/endeca/PlatformServices/11.1.0/tools/server/bin/

Logs are written to /usr/local/endeca/PlatformServices/workspace/logs


[vagrant@localhost software]$ /usr/local/endeca/ToolsAndFrameworks/11.1.0/server/bin/

Logs are written to /usr/local/endeca/ToolsAndFrameworks/11.1.0/server/workspace/logs


[vagrant@localhost software]$ /usr/local/endeca/CAS/11.1.0/bin/ &

Logs are written to /usr/local/endeca/CAS/workspace/logs

Install ATG

Install ATG Platform


[vagrant@localhost software]$ ./OCPlatform11.1.bin

Use the following locations:

  • JBoss: /home/vagrant/jboss
  • Java: /usr/java/jdk1.7.0_67

Install ATG CRS


[vagrant@localhost software]$ ./OCReferenceStore11.1.bin

Configure ATG:


Follow the CRS instructions, except as noted here. Begin with the "Starting CIM" section of the CRS setup guide.

Exceptions to default values:

  • If you've installed your database using these instructions, its IP address will be, port 1521, SID orcl
  • Very Important: The jps-config.xml that you specify for the EAC application MUST be the one from the ToolsAndFrameworks install and NOT the one from the ATG install. The ATG install one will give you errors when you try to create the app. This is the correct jps-config.xml file: /usr/local/endeca/ToolsAndFrameworks/11.1.0/deployment_template/lib/../../server/workspace/credential_store/jps-config.xml

You can skip the "Reducing Logging Messages" and "Additional Application Server Configuration" steps. Continue through "Accessing the Storefront"


There are a couple of settings that are absolutely critical to getting this cluster running, but which aren't obvious and are difficult to diagnose if things go wrong. Don't worry, both of these should be taken care of automatically in the scripts.

  • You must have -Duser.timezone=UTC in your JAVA_OPTS variable that's used to run jboss and cim. For some reason it's required when connecting from the atg vm to the db vm. If you don't have this you'll get user timezone errors and ultimately can't use jdbc to connect to your database.
  • You must use the jps-config.xml file that comes with the ToolsAndFrameworks' deployment template: /usr/local/endeca/ToolsAndFrameworks/11.1.0/deployment_template/lib/../../server/workspace/credential_store/jps-config.xml If you're not sure if you set it right when configuring your application in cim, check in the file /usr/local/endeca/Apps/CRS/config/script/WorkbenchConfig.xml. If you don't use the correct jps-config.xml, you'll get a variety of strange and unhelpful errors, like a ClassNotFoundException when removing your application with