24.7. Clustered Instances

Procedure 24.12. Launch Clustered JBoss EAP 6 AMIs

1. Select an AMI.
2. Define the desired number of instances (the cluster size).
3. Select the VPC and instance type.
4. Click on Security Group.
5. Ensure that all traffic from the JBoss EAP 6 cluster subnet is allowed.
6. Define other restrictions as desired.
7. Add the following into the User Data field:

   Example 24.7. Example User Data Field

   ## mod cluster proxy addresses
   MOD_CLUSTER_PROXY_LIST=10.0.0.4:7654
   
   ## clustering setup
   JBOSS_JGROUPS_S3_PING_SECRET_ACCESS_KEY=<your secret key>
   JBOSS_JGROUPS_S3_PING_ACCESS_KEY=<your access key>
   JBOSS_JGROUPS_S3_PING_BUCKET=<your bucket name>
   
   ## password to access admin console
   JBOSSAS_ADMIN_PASSWORD=<your password for opening admin console>
   
   ## database credentials configuration
   JAVA_OPTS="$JAVA_OPTS -Ddb.host=instancename.something.rds.amazonaws.com -Ddb.database=mydatabase -Ddb.user=<user> -Ddb.passwd=<pass>"
   
   ## subnet prefix this machine is connected to
   SUBNET=10.0.1.
   
   #### to run the example no modifications below should be needed ####
   PORTS_ALLOWED="1024:65535"
   JBOSS_IP=`hostname | sed -e 's/ip-//' -e 'y/-/./'` #listen on public/private EC2 IP address
   
   cat > $USER_SCRIPT << "EOF"
   ## Get the application to be deployed from an Internet URL
   # mkdir -p /usr/share/java/jboss-ec2-eap-applications
   # wget https://<your secure storage hostname>/<path>/<app name>.war -O /usr/share/java/jboss-ec2-eap-applications/<app name>.war
   
   ## install the JDBC driver as a core module
   yum -y install mysql-connector-java
   mkdir -p /usr/share/jbossas/modules/com/mysql/main
   cp -v /usr/share/java/mysql-connector-java-*.jar /usr/share/jbossas/modules/com/mysql/main/mysql-connector-java.jar
   
   cat > /usr/share/jbossas/modules/com/mysql/main/module.xml <<"EOM"
   <?xml version="1.0" encoding="UTF-8"?>
   <module xmlns="urn:jboss:module:1.0" name="com.mysql">
      <resources>
         <resource-root path="mysql-connector-java.jar"/>
      </resources>
      <dependencies>
         <module name="javax.api"/>
      </dependencies>
   </module>
   EOM
   
   cat > $USER_CLI_COMMANDS << "EOC" 
   ## Deploy sample application from local filesystem
   deploy --force /usr/share/java/jboss-ec2-eap-samples/cluster-demo.war
   
   ## ExampleDS configuration for MySQL database
   data-source remove --name=ExampleDS
   /subsystem=datasources/jdbc-driver=mysql:add(driver-name="mysql",driver-module-name="com.mysql")
   data-source add --name=ExampleDS --connection-url="jdbc:mysql://${db.host}:3306/${db.database}" --jndi-name=java:jboss/datasources/ExampleDS --driver-name=mysql --user-name="${db.user}" --password="${db.passwd}"
   /subsystem=datasources/data-source=ExampleDS:enable
   /subsystem=datasources/data-source=ExampleDS:test-connection-in-pool
   EOC
   
   ## this will workaround the problem that in a VPC, instance hostnames are not resolvable
   echo -e "127.0.0.1\tlocalhost.localdomain localhost" > /etc/hosts
   echo -e "::1\tlocalhost6.localdomain6 localhost6" >> /etc/hosts
   for (( i=1 ; i<255 ; i++ )); do
      echo -e "$SUBNET$i\tip-${SUBNET//./-}$i" ;
   done >> /etc/hosts
   
   EOF

   Copy to Clipboard Toggle word wrap

Procedure 24.13. Testing the Clustered Instance

1. Navigate to http://ELASTIC_IP_OF_APACHE_HTTPD in a browser to confirm the web server is running successfully.

2. Test the Clustered Nodes

   1. Navigate to http://ELASTIC_IP_OF_APACHE_HTTPD/cluster-demo/put.jsp in a browser.
   2. Verify that one of the cluster nodes logs the following message:

      Putting date now

      Copy to Clipboard Toggle word wrap
   3. Stop the cluster node that logged the message in the previous step.
   4. Navigate to http://ELASTIC_IP_OF_APACHE_HTTPD/cluster-demo/get.jsp in a browser.
   5. Verify that the time shown is the same as the time that was PUT using put.jsp in Step 2-a.
   6. Verify that one of the running cluster nodes logs the following message:

      Getting date now

      Copy to Clipboard Toggle word wrap
   7. Restart the stopped clustered node.
   8. Connect to the Apache HTTP server instance:

      ssh -L7654:localhost:7654 <ELASTIC_IP_OF_APACHE_HTTPD>

      Copy to Clipboard Toggle word wrap
   9. Navigate to http://localhost:7654/mod_cluster-manager to confirm all instances are running correctly.