IBP Development Utilities

Here's how you create a working BMS environment on your local Linux/Mac machine from scratch.

Prerequisite Software

If you do not have any of the following installed, follow the instructions.

MySQL

Download and install the latest version of the community edition.

You will need to add mysql to your path. When you restart your computer, the MySQL daemon should be running. To start it immediately, you can run sudo /Library/StartupItems/MySQLCOM/MySQLCOM start.

You will probably also want to install MySQL Workbench and MySQL Utilities to assist in working with your MySQL databases.

Tomcat

Download and extract version 7.0.50 of Tomcat.

NOTE: It is important that you do not use a version more recent than 7.0.50, due to an issue with load time weaving in 7.0.51+.

JDK

Install the latest Java 7.x JDK and set the JAVA_HOME variable.

Maven

Install the latest version of Maven and follow the installation instructions.

Groovy

Install the latest version of Groovy and follow the installation instructions.

Project Code

Checkout all BMS projects from Git:

• git clone https://github.com/digitalabs/IBPMiddleware
• git clone https://github.com/digitalabs/IBPCommons
• git clone https://github.com/digitalabs/GermplasmStudyBrowser
• git clone https://github.com/digitalabs/BreedingManager
• git clone https://github.com/digitalabs/DatasetImporter
• git clone https://github.com/digitalabs/Fieldbook
• git clone https://github.com/digitalabs/IBPWorkbench

Configuring Tomcat

Load Time Weaving

This is one off Tomcat setting required to deal with Spring Load Time Weaving used by IBP workbench.

• Download spring-instrument-3.1.1.RELEASE.jar and save it to a known location.
• Alter the -javaagent property in the CATALINA_OPTS variable in tomcat_settings/setenv.sh to be the location of spring-instrument-3.1.1.RELEASE.jar.
• Copy the tomcat_settings/setenv.sh file into your TOMCAT_HOME/bin directory (this file probably does not already exist but if it does, add to it). Tomcat will pick this file up (check in catalina.sh if you like) when it starts.

Tomcat Port

• Popular/default tomcat port used by IBP applications is 18080. Configure the default tomcat Connector in TOMCAT_HOME/conf/server.xml to listen on 18080 instead of the default 8080.

Create DB Schema

The groovy script DROPS YOUR EXISTING DB COMPLETELY so backup first, if you need to.

• Configure config.properties to point to your MySQL instance
• Run groovy createschema.groovy. To see Grapes being downloaded (useful for first time running), add verbose Grape handling to your command groovy -Dgroovy.grape.report.downloads=true createschema.groovy

The groovy script:

• Drops and creates the workbench schema
• Registers Crops and Tools in workbench schema
• Drops and creates central and local database schema for one sample crop e.g. ibdbv2_rice_central and ibdbv2_rice_local
• Applies the update scripts that create views, procedures etc. in workbench, local and central schema

Build the Projects

• Project configuration can be found in {project-folder}/pipeline/config. You can use an existing folder to start with, or create a new folder with your own configuration. At a minimum, ensure any config you are using has the correct settings to connect to your local MySQL database.
• Build each project by running mvn clean antrun:run install -DenvConfig=[config_folder], where config_folder is the folder above. Be aware that you need to build Middleware first, and Commons second (as it depends on Middleware). The rest can be built in any order - each depending only on Commons.

Note: There is an issue with the tests running at the moment - please add -DskipTests to skip the tests and allow the build to pass.

Load Crop Data

The crop data files are often large, so we do not keep these in Git. Follow these directions to load:

1. Download the crop installer exe from the Efficio GCP confluence. Rename the extension from .exe to .zip and use the command line tool to unzip, i.e. unzip installer.zip -d destination_folder.
2. Open script database/crop/rice/loadRice.sh for editing
   • Set DATA\_SCRIPT\_LOCATION to refer to the directory where you extracted the crop data dump sql files.
   • Set USER, PASSWORD, PORT and CENTRAL_CROP_DB values as per your local MySQL instance setup.
3. Setting max_allowed_packet=64M in MySQL(d) configuration (usually /etc/my.cnf) will be helpful. Otherwise you may see ERROR 2006 (HY000) at line 17: MySQL server has gone away when loading some large scripts, which is not good. If this is the first item in your configuration file, be aware you need to include it in a group. If this is the only thing in your file, you can use:

[mysqld]
max_allowed_packet	= 50M

4. cd database/crop/rice
5. ./loadRice.sql

Deploy Wars to Tomcat

Configure the scripts located in deploy_scripts to reflect your system configuration (tomcat and checkout location). Be aware that the checkout location is the root directory where all your project folders are.

Each script in the deploy_scripts directory does the following:

• stops Tomcat (doesn't matter if it is already stopped)
• sleeps for 15 seconds to make sure it is stopped
• copies the war(s)
• starts tomcat

Which Script to Run

• deploy_all.sh: Stops Tomcat - copies 5 war files into tomcat/webapps. Please configure your Tomcat home dir (TOMCAT_SERVER), and the directory where all of your checkouts reside (CHECKOUT_HOME), we copy from the Maven target dir of each project.
• deploy_app.sh: For Single apps APART FROM Workbench. Configure as above. Example usage is ./deploy_app.sh Fieldbook.
• deploy_wb.sh: Deploys workbench. No parameters required.

Usually you'll just want to run ./deploy_all.sh.

Errors on copy will be written to console. Watch the Tomcat logs for progress.