The Next Milestone

HBase Spring Data Sample

2015-07-25T09:10:00.000-07:00

Document Version 1.0

Keywords:
Spring Data for Hadoop, NoSQL, HBase, Java, Maven

1. Create a simple Maven project "spring-hbase" in Eclipse

2. Prepare project's pom.xml file

3. Prepare Spring configuration file

4. Write code accessing HBase

5. Run the sample

N oSQL database might be interesting even if we are not really dealing with petabytes data. For a relational database with terabytes data, the time a query need to finish might take so long that it becomes NOT acceptable for a performance critical system. Even if we index tables, we optimize queries, or do VACUUM every day. For one of my concrete problems, I tried a solution with HBase and Spring Data for Hadoop, with success, and with far better performance. I had problems making my first Spring Data application to run. This post will try to show a simple running sample of Spring Data for Hadoop. Hope this could help someone with more or less the same difficulties.

1. Create a simple Maven project "spring-hbase" in Eclipse

2. Prepare project's pom.xml file

Necessary dependencies include jars of Spring, Spring Data, HBase and Hadoop.
In case some jars could not be found in maven central repositories, you could install them manually in enterprise repositories or you local repository.

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<groupId>wangs</groupId>
<artifactId>spring-hbase</artifactId>
<version>0.0.1-SNAPSHOT</version>
<name>spring-hbase</name>
<repositories>
<repository>
<id>repo1.maven.org/maven2</id>
<url>http://repo1.maven.org/maven2</url>
</repository>
</repositories>
<dependencies>
<dependency>
<groupId>org.springframework.data</groupId>
<artifactId>spring-data-hadoop</artifactId>
<version>2.2.0.RELEASE</version>
</dependency>
<dependency>
<groupId>org.springframework</groupId>
<artifactId>spring-core</artifactId>
<version>4.1.7.RELEASE</version>
</dependency>
<dependency>
<groupId>org.springframework</groupId>
<artifactId>spring-context</artifactId>
<version>4.1.7.RELEASE</version>
</dependency>
<dependency>
<groupId>org.apache.hbase</groupId>
<artifactId>hbase-client</artifactId>
<version>1.0.1.1</version>
</dependency>
<dependency>
<groupId>org.apache.hbase</groupId>
<artifactId>hbase-common</artifactId>
<version>1.0.1.1</version>
</dependency>
<dependency>
<groupId>org.apache.hbase</groupId>
<artifactId>hbase-hadoop2-compat</artifactId>
<version>1.0.1.1</version>
</dependency>
<dependency>
<groupId>org.apache.hbase</groupId>
<artifactId>hbase-protocol</artifactId>
<version>1.0.1.1</version>
</dependency>
<dependency>
<groupId>org.apache.htrace</groupId>
<artifactId>htrace-core</artifactId>
<version>3.1.0-incubating</version>
</dependency>
<dependency>
<groupId>io.netty</groupId>
<artifactId>netty-all</artifactId>
<version>4.0.23.Final</version>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-client</artifactId>
<version>2.5.1</version>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-common</artifactId>
<version>2.5.1</version>
</dependency>
<dependency>
<groupId>org.apache.hadoop</groupId>
<artifactId>hadoop-hdfs</artifactId>
<version>2.5.1</version>
</dependency>
<dependency>
<groupId>log4j</groupId>
<artifactId>log4j</artifactId>
<version>1.2.17</version>
</dependency>
</dependencies>
</project>

3. Prepare Spring Configuration file

With Spring Data for Hadoop, the connection to HBase is managed by Spring, this include:

Hadoop configuration for HBase and
a HbaseTemplate, the unified access to HBase

So we create a spring configuration file "SpringBean.xml" with the content below:

<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns:context="http://www.springframework.org/schema/context"
xmlns:hdp="http://www.springframework.org/schema/hadoop"
xmlns:p="http://www.springframework.org/schema/p"
xsi:schemaLocation="http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans.xsdxmlns
http://www.springframework.org/schema/context http://www.springframework.org/schema/context/spring-context.xsd
http://www.springframework.org/schema/hadoop http://www.springframework.org/schema/hadoop/spring-hadoop.xsd">

<context:component-scan base-package="springdt" />
<context:property-placeholder location="hbase.properties"/>

<hdp:configuration id="hadoopConfiguration">
fs.defaultFS=hdfs://127.0.0.1:9000
</hdp:configuration>

<hdp:hbase-configuration configuration-ref="hadoopConfiguration" zk-quorum="127.0.0.1" zk-port="2181"/>

<bean id="hbaseTemplate" class="org.springframework.data.hadoop.hbase.HbaseTemplate">
<property name="configuration" ref="hbaseConfiguration"/>
</bean>
<bean id="hBaseService" class="springdt.HBaseService"/>

</beans>

The "HbaseService" bean is the only bean of this sample. We will show the code right away.

4. Write code accessing HBase

In this sample we will create a table in HBase called "report", and try store some data in the table. The table has one column family called "data", and only one column named "file". The value of "file" is file name, which could be report name in real life.

package springdt;

import javax.inject.Inject;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.hbase.HColumnDescriptor;
import org.apache.hadoop.hbase.HTableDescriptor;
import org.apache.hadoop.hbase.client.HBaseAdmin;
import org.apache.hadoop.hbase.client.HTableInterface;
import org.apache.hadoop.hbase.client.Put;
import org.apache.hadoop.hbase.util.Bytes;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.context.support.AbstractApplicationContext;
import org.springframework.context.support.ClassPathXmlApplicationContext;
import org.springframework.data.hadoop.hbase.HbaseTemplate;
import org.springframework.data.hadoop.hbase.TableCallback;
import org.springframework.stereotype.Service;

/**
* Demonstrating how to access HBase using Spring Data for Hadoop.
*
* @author swang
*
*/
@Service
public class HBaseService {

@Autowired
private Configuration hbaseConfiguration;

@Inject
private HbaseTemplate hbTemplate;

// Table info
final String tableName = "report";
final String columnFamilyData = "data";
final String colFile = "file";
final String rowNamePattern = "row";
final String value = "report24.csv-";

/**
*
* @throws Exception
*/

public void run() throws Exception {
// 1. create table
createTable();
// 2. add data entry
addData();
}

/**
* Creates HBase table
*
* @throws Exception
*/
public void createTable() throws Exception {
HBaseAdmin admin = new HBaseAdmin(hbaseConfiguration);

if (admin.tableExists(tableName)) {
admin.disableTable(tableName);
admin.deleteTable(tableName);
}

HTableDescriptor tableDes = new HTableDescriptor(tableName);
HColumnDescriptor cf1 = new HColumnDescriptor(columnFamilyData);
tableDes.addFamily(cf1);
admin.createTable(tableDes);
}

/**
* Adds data entry for report.
*/
private void addData() {
hbTemplate.execute(tableName, new TableCallback<Boolean>() {

public Boolean doInTable(HTableInterface table) throws Throwable {
for (int i = 0; i < 1000; i++) {
Put p = new Put(Bytes.toBytes(rowNamePattern + i));
p.add(Bytes.toBytes(columnFamilyData),
Bytes.toBytes(colFile), Bytes.toBytes(value + i));
table.put(p);
}
return new Boolean(true);
}
});
}

public static void main(String[] args) throws Exception {
AbstractApplicationContext ctx = new ClassPathXmlApplicationContext(
"SpringBeans.xml");

HBaseService hBaseService = (HBaseService) ctx.getBean("hBaseService");
hBaseService.run();
}
}

The "SpringBeans.xml" we created in step 3 should be put in "src/main/resources", so that it could be found at runtime.

5. Run the sample

We assume you have HBase over Hadoop installed. Make sure the HBase version is compatible with Hadoop. In my case I have "hbase-1.0.1.1" and "hadoop-2.7.0".
Now start HBase server.
Run "HBaseService" in Eclipse as java application. In Eclipse console you will see output like this:

2015-07-25 15:10:52 INFO ClientCnxn:852 - Socket connection established to 127.0.0.1/127.0.0.1:2181, initiating session
2015-07-25 15:10:52 INFO ClientCnxn:1235 - Session establishment complete on server 127.0.0.1/127.0.0.1:2181, sessionid = 0x14ec4b8cf30000e, negotiated timeout = 90000
2015-07-25 15:10:53 INFO HBaseAdmin:978 - Started disable of report
2015-07-25 15:10:55 INFO HBaseAdmin:1033 - Disabled report
2015-07-25 15:10:55 INFO HBaseAdmin:738 - Deleted report

"HBaseService" should have created 1000 ehtries in HBase table "report". Check this with hbase shell:

hbase(main):003:0> scan "report"
ROW COLUMN+CELL
row0 column=data:file, timestamp=1437823593865, value=report24.csv-0
row1 column=data:file, timestamp=1437823593897, value=report24.csv-1
...
row999 column=data:file, timestamp=1437823603565, value=report24.csv-999
1000 row(s) in 2.0600 seconds

Now we are done with the simple example. Have fun!

ActiveMQ Configuration for persistent Message Delivery

2013-09-13T00:14:00.000-07:00

Document Version 1.0

Copyright © 2012-2013 beijing.beijing.012@gmail.com

Wether message is delivered persistent or not, depends on both message producer. i.e. the client, and configuration of server, in our case ActiveMQ.

1. On the Client Side

Message Producer can tell ActiveMQ to deliver message in persistent mode in 2 ways:

- with MessageProducer#setDeliveryMode(..), this set the default delivery mode for all messges sent with this producer.

- MessageProducer# send(..), this set delivery mode on message level, i.e. set delivery mode for each message.

2. On the Server Side

If persistence delivery is not enabled on ActiveMQ, messages will not be delivered in persistence mode even if message producer set delivery mode as "PERSISTENT"

2.1 Set persistent mode

With ActiveMQ, persistent node is set on broker tag of activemq.xml

The default:

When no persistent properties is set for broker, the default is persistent="true". i.e.

2.2 Configure persistence Adapter

When broker is configured to run in persistence mode, it requires a "persistenceAdapter", for example:

</persistenceAdapter>

The above is also the default configuration, if no persistence apdapter is configured in activemq.xml

2.3 Pesistence Store Usage

Having configured persistence adapter as "khdb", as next step we need to tell how much disk size is available for "khdb" to store messages:

</memoryUsage>

<storeUsage>

<storeUsage limit="100 gb"/>

</storeUsage>

</tempUsage>

</systemUsage>

When no "systemUsage" section is configured, the ActiveMQ use default 100gb for "khdb" storeUsage.

When there is no enough space available on the disk, ActiveMQ will show ERROR/WARN message at startup like:

2013-09-13 09:05:15,136 | WARN | Store limit is 102400 mb, whilst the data directory: /activemq/data/localhost/KahaDB only has 20603 mb of usable space | org.apache.activemq.broker.BrokerService | main

ActiveMQ can be started and live with such error/warnings, But if you are running a production system, be sure to check if you really need more disk.

2.4 Running in "NON_PERSISTENT" mode

When broker is configured with "persistent=false", then ActiveMQ persists no messges at all in DB even if client set delivery mode as "persistent".

Running in "persistent=false" mode, ActiveMQ use default MemoryPersistenceAdapter. All messages resides in ActiveMQ' system memory. When the limit of the system usage is reached, message will be stored in "tempStore" on file system.

<memoryUsage>

<memoryUsage limit="10 gb"/>

</memoryUsage>

</storeUsage>

<tempUsage>

<tempUsage limit="50 gb"/> 

</tempUsage>

</systemUsage>

Here the "memoryUsage" of system memory, is RAM, and is not the whole memory used by ActiveMQ, but only the memory used for storing messages. When the total available RAM on the machine is less than what is configured above, ActiveMQ can still start and works, but logs error info like:

2013-09-13 09:49:19,078 | ERROR | Memory Usage for the Broker (10240 mb) is more than the maximum available for the JVM: 981 mb | org.apache.activemq.broker.BrokerService | main

My test machine has only 8gb RAM, but here above configuration requires 10gb for ActiveMQ's in memory store.

Beside the above error, at sartup, my ActiveMQ show also following error info:

2013-09-13 09:49:19,078 | ERROR | Temporary Store limit is 51200 mb, whilst the temporary data directory: /activemq/data/localhost/tmp_storage only has 20570 mb of usable space |

This is because, ActiveMQ is configured to use 50GB temp space on disk, howeve my machine has only about 20gb free.

The configuration for "storeUsage" here is not used by ActiveMQ, sicnce in non-persistent mode, no data is writen is written to DB.

HTML5 WebSocket Sample with Java Servlet on Server Side

2013-07-15T00:16:00.003-07:00

Document Version 1.0

Copyright © 2012-2013 beijing.beijing.012@gmail.com

Beside Server-Sent Events, WebSocket is an another important feature introduced by HTML5 for browser to update web page content automatically from server. The most important differences between WebSocket and Server-Sent Events are:

Server-Sent Events is one way communication whereas WebSocket use full-duplex channel.
Server-Sent Event is HTTP, WebSocket introduced new protocol "ws" protocol

Here I am not going compare WebSocket and Server-Sent Events in detail, I will just show you a simple WebSecket sample to get a feeling.

A sample of Server-Sent Events could be found here http://milestonenext.blogspot.de/2013/07/html5-server-sent-events-sample-with.html

In the example below, browser opens a window, send text message to server. Server just send message back to browser, shown in the window. (by the way, the sample could be extended with small effort to work as a chat application ...)

The client is just a html page with embedded javascript. The Server is a Java Servlet web application running on Tomcat 7.0.29. (WebSocket, "ws" protocol support with Tomcat since version 7.0.24 ).

1. Create a simple web application "WebSocketSample"

User your favorite IDE (I use Eclipse) to create a Web Application. Be sure to use servlet 3.x style.

The sample web application consists of :

two java classes, which are the server side code
one "html" page which is the client

2. MyWebSocketServlet.java

The sample uses Tomcat implementation of of "WebSocket", so we need to have "catalina.jar" on class path.
Add servlet-api.jar to class path as well.

package sample;

import java.util.concurrent.ConcurrentHashMap;

import javax.servlet.annotation.WebServlet;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpSession;

import org.apache.catalina.websocket.StreamInbound;
import org.apache.catalina.websocket.WebSocketServlet;

/**
* WebSocketServlet is contained in catalina.jar. It also needs servlet-api.jar
* on build path
*
* @author wangs
*
*/
@WebServlet("/wsocket")
public class MyWebSocketServlet extends WebSocketServlet {

private static final long serialVersionUID = 1L;

// for new clients, <sessionId, streamInBound>
private static ConcurrentHashMap<String, StreamInbound> clients = new ConcurrentHashMap<String, StreamInbound>();

@Override
protected StreamInbound createWebSocketInbound(String protocol,
HttpServletRequest httpServletRequest) {

// Check if exists
HttpSession session = httpServletRequest.getSession();

// find client
StreamInbound client = clients.get(session.getId());
if (null != client) {
return client;

} else {
client = new MyInBound(httpServletRequest);
clients.put(session.getId(), client);
}

return client;
}

public StreamInbound getClient(String sessionId) {
return clients.get(sessionId);
}

public void addClient(String sessionId, StreamInbound streamInBound) {
clients.put(sessionId, streamInBound);
}
}

3. MyInbund.java

Since Tomcat's WsOutbound has indirect dependency on "tomcat-koyote.jar", please add this jar to class path to resolve compile error like "the hierarchy of the type ... is inconsistent ..."

package sample;

import java.io.IOException;
import java.nio.ByteBuffer;
import java.nio.CharBuffer;

import javax.servlet.http.HttpServletRequest;

import org.apache.catalina.websocket.MessageInbound;
import org.apache.catalina.websocket.WsOutbound;

/**
* Need tomcat-koyote.jar on class path, otherwise has compile error "the hierarchy of the type ... is inconsistent"
* @author wangs
*
*/
public class MyInBound extends MessageInbound{

private String name;

private WsOutbound myoutbound;

public MyInBound(HttpServletRequest httpSerbletRequest) {

}
@Override
public void onOpen(WsOutbound outbound) {
System.out.println("on open..");
this.myoutbound = outbound;
try {
this.myoutbound.writeTextMessage(CharBuffer.wrap("hi, what's your name?"));

} catch (Exception e) {
throw new RuntimeException(e);
}

}

@Override
public void onClose(int status) {
System.out.println("Close client");
//remove from list
}

@Override
protected void onBinaryMessage(ByteBuffer arg0) throws IOException {

}

@Override
protected void onTextMessage(CharBuffer inChar) throws IOException {

System.out.println("Accept msg");
CharBuffer outbuf = CharBuffer.wrap("- " + this.name + " says : ");
CharBuffer buf = CharBuffer.wrap(inChar);

if(name != null) {
this.myoutbound.writeTextMessage(outbuf);
this.myoutbound.writeTextMessage(buf);
} else {
this.name = inChar.toString();

CharBuffer welcome = CharBuffer.wrap("== Welcome " + this.name + "!");
this.myoutbound.writeTextMessage(welcome);
}

this.myoutbound.flush();

}

}

4. wssockrt.html

<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<title>Tomcat web socket</title>
<script type="text/javascript">
var ws = new WebSocket("ws://localhost:8080/WebSocketSample/wsocket");
ws.onopen = function () {
};

ws.onmessage = function(message) {
document.getElementById("msgArea").textContent += message.data + "\n";
};

function postToServer() {
ws.send(document.getElementById("msg").value);
document.getElementById("msg").value = "";
}

function closeConnect() {
ws.close();
}
</script>
</head>

<body>
<div>
<textarea rows="4" cols="100" id="msgArea" readonly></textarea>
</div>
<div>
<input id="msg" type="text"/>
<button type="submit" id="sendButton" onclick="postToServer()">Send</button>
</div>
</body>
</html>

5. Export ".war" and deploy it on a running Tomcat

Export the "WebSocketSample.war"and deploy it on a running Tomcat server.

Open the link below with browser:

http://localhost:8080/WebSocketSample/wsocket.html

You will see:

Now type your name and send:

So it woks!

6. The "WebSocket" Header

When we take a close look at the headers of the Request and Response above:

We will see, browser is not speaking "http" with server any more, now it speaks "websocket" protocol.

HTML5 Server-Sent Events Sample with Java Servlet as Event Server

2013-07-09T13:06:00.001-07:00

Document Version 1.0

Copyright © 2012-2013 beijing.beijing.012@gmail.com

Server-Sent Events is a HTML5 feature which support browse to update page content automatically.

I will show a Server-Sent Events sample which get update of serve time automatically. Server side implementation is a Java Servlet running on Tomcat.

1. Create a simple web application:

2. Create a 3.X style Java Servlet "SseServer.java"

package sse;

import java.io.IOException;
import java.io.PrintWriter;
import java.util.Date;

import javax.servlet.ServletException;
import javax.servlet.annotation.WebServlet;
import javax.servlet.http.HttpServlet;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;

/**
*
* @author swang
*
*/
@WebServlet("/SseServer")
public class SseServer extends HttpServlet {
private static final long serialVersionUID = 1L;

protected void doGet(HttpServletRequest request,
HttpServletResponse response) throws ServletException, IOException {
doPost(request, response);
}

protected void doPost(HttpServletRequest request,
HttpServletResponse response) throws ServletException, IOException {
// Besides "text/event-stream;", Chrome also needs charset, otherwise
// does not work
// "text/event-stream;charset=UTF-8"
response.setContentType("text/event-stream;charset=UTF-8");
response.setHeader("Cache-Control", "no-cache");
response.setHeader("Connection", "keep-alive");

PrintWriter out = response.getWriter();

while (true) {
out.print("id: " + "ServerTime" + "\n");
out.print("data: " + new Date().toLocaleString() + "\n\n");
out.flush();
// out.close(); //Do not close the writer!
try {
Thread.currentThread().sleep(5000);
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}

}

Make sure in the Servlet, set content type of response as "text/event-stream;charset=UTF-8", and NOT
just "text/event-stream". Otherwise it does not work with Chrome.

3. Create a SSE.html file

<!DOCTYPE html>
<html>
<body>
<h1>Current Server Time : </h1>

<div id="ServerTime"></div>

<script>
if (typeof (EventSource) !== "undefined") {
var source = new EventSource("http://localhost:8080/SSE/SseServer");
source.onmessage = function(event) {
document.getElementById("ServerTime").innerHTML += event.data
+ "<br><br>";
};
} else {
document.getElementById("ServerTime").innerHTML = "Sorry, your browser does not support server-sent events...";
}
</script>

</body>
</html>

4. Deploy the Web Application

Export the Web Application as "SSE.war", and deploy it to a running Tomcat Server.

5. Try accessing the SSE.html use browser

Access the SSE.html using link http://localhost:8080/SSE/SSE.html

Your browser should shown content as below:

When every 5 seconds, there is new line shown in your browser, congratulations, you have made your first Server-Sent Events Application!

JMX Sample: monitor standalone Application use own JMX Server

2013-07-06T22:36:00.004-07:00

Document Version 1.0

Copyright © 2012-2013 beijing.beijing.012@gmail.com

When you are familiar with JBoss, you should have known JMX-console. When you want to monitor you application deployed on JBoss, you might go with following step:

wrapp the to be monitored property in a Mbean,
register the MBean with JBoss' JMX sesrver
check / manage the properties through JMX-Console

But what if you want to monitor a standalone application, and there is no JMX-server to use?

In such case, your will have to start your own JMX-server.

I will show how to use your own JMX-server, to monitor a standalone application which monitors the application's local time.

1. The MBean

The MBean interface exposes which methods are monitored by JMX.

package mbean;

/**
*
* @author swang
*
*/
public interface MySampleMBean {
String getServerTime();
}

This is the MBean implementation. The property "serverTime" will be updated every 5 seconds. We expect the upated serverTime will be shown to a JMX client when it connects to JXX-Server.

package mbean;

import java.util.Date;

/**
*
* @author swang
*
*/
public class MySample implements MySampleMBean, Runnable{

private static MySample instance = new MySample();
private String serverTime;

private MySample() {
}

public static MySample getInstance() {
return instance;
}

@Override
public String getServerTime() {
return this.serverTime;
}

public void setServerTime(String serverTime) {
this.serverTime = serverTime;
}

@Override
public void run() {
while(true) {
//Updating serverTime every 5 seconds
setServerTime(new Date().toLocaleString());

try {
Thread.currentThread().sleep(5000);
} catch (InterruptedException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
}
}

2. A JMX Agent

This is the main class of the program, we call it a JMX agent. It starts a JMX-Server, starts the MBean, and regist the MBean with the JMX-Server.

package mbean;

import java.lang.management.ManagementFactory;
import java.rmi.registry.LocateRegistry;
import java.util.HashMap;

import javax.management.MBeanServer;
import javax.management.ObjectName;
import javax.management.remote.JMXConnectorServer;
import javax.management.remote.JMXConnectorServerFactory;
import javax.management.remote.JMXServiceURL;

/**
*
* @author swang
*
*/
public class MyJMXAgent {

private static MBeanServer mbs = null;

public void init() throws Exception {
mbs = ManagementFactory.getPlatformMBeanServer();

ObjectName objName = new ObjectName("TestJMXAgent:name=MySampleMBean");
mbs.registerMBean(MySample.getInstance(), objName);

String hostname = "localhost";
int port = 1717;
LocateRegistry.createRegistry(port);
System.setProperty("java.rmi.server.randomIDs", "true");
System.setProperty("java.rmi.server.hostname", hostname);

HashMap<String, Object> env = new HashMap<String, Object>(3);
env.put("jmx.remote.x.password.file", "jmxremote.password");
env.put("jmx.remote.x.access.file", "jmxremote.access");

JMXServiceURL url = new JMXServiceURL("service:jmx:rmi://" + hostname
+ ":" + port + "/jndi/rmi://" + hostname + ":" + port
+ "/jmxrmi");
JMXConnectorServer cs = JMXConnectorServerFactory
.newJMXConnectorServer(url, env, mbs);

cs.start();

System.out.println("Starting JMXConnectorServer on: " + url);
}

public static void main(String[] args) throws Exception {
System.out.println("jmx agent starting");
// Start updating MBean
new Thread(MySample.getInstance()).start();

MyJMXAgent agent = new MyJMXAgent();
agent.init();

}
}

3. Configuration JMX access:

Create a "jmxremote.access" file:

monitor readonly

Create a "jmxremote.password" file:

monitor password

4. Run MyJMXAgent as Java application:

jmx agent starting
host name: localhost
Starting JMXConnectorServer on: service:jmx:rmi://localhost:1717/jndi/rmi://localhost:1717/jmxrmi

A JMX-Server is now running on localhost:1717, and you could use a JMX-client to connect to it.

5. Connect to JMX-Server with jconsole

Start jconsole an connect to MyMXAgent:

Check monitored property "serverTime":

Click "Refresh" button to get the updated "serverTime" :

Create Document with Digital Signature using GPG on Linux

2013-07-02T13:44:00.000-07:00

Document Version 1.0

Copyright © 2012-2013 beijing.beijing.012@gmail.com

My friend Tomy has got an offer from GooPle, greate, isn't it? GooPle! On the Phone, the HR manager, Mr. Bozz of GooPle says, he will sign the contract and send it to Tomy right away as post. But at the same time, Mr. Bozz also pointed out, that Tomy should sign the contract and send it back to GooPle in 8 hours. 8 hours? Tomy is even not sure if he could receive the post in 8 hours, since he is living in a small village, the post man comes only once a week...

Of course Tomy do not want to miss this chance, by GooPle. "Sorry sir, but could you sign it and send me as file, and I will sign it with digital signature, so you would get the signed contract in 20 minutes "
"Yes, of course ..."

In the following section, I will show step by step, how does Mr. Bozz prepare the contract, sign it with his digital signature, and how will Tomy verify the signature, and extract the original content of the contract, otherwise the content is encrypted and not readable.

Step 1. Mr. Bozz prepare the original contract, and save the file as "contract.txt" :

"This is a contract between Mr. Tomy and GooPle ..."

Step 2. Mr. Bozz sign the "contract.txt" file with his digital signature

Mr. Bozz's computer is a Linux (Mint 14) machine, he knows there is a tool under linux called "GPG (GNU Privacy Guard)", which could be used to create digital signature.

He opens a terminal and types following command:

bozz@PC007x /home/bozz $ gpg --sign contract.txt

Console output shows:

gpg: directory `/home/tomy/.gnupg' created
gpg: new configuration file `/home/tomy/.gnupg/gpg.conf' created
gpg: WARNING: options in `/home/tomy/.gnupg/gpg.conf' are not yet active during this run
gpg: keyring `/home/tomy/.gnupg/secring.gpg' created
gpg: keyring `/home/tomy/.gnupg/pubring.gpg' created
gpg: no default secret key: secret key not available
gpg: signing failed: secret key not available

GPG complains here it can not find the private and public key of Mr. Bozz, but these keys are need for creating digital signature.

Step3. Mr. Bozz creates his private and public key:

To generate keys, Mr. Bozz types following command in the terminal:

bozz@PC007x /home/bozz $ gpg --gen-key

Terminal output:

gpg (GnuPG) 1.4.11; Copyright (C) 2010 Free Software Foundation, Inc.
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

Please select what kind of key you want:
(1) RSA and RSA (default)
(2) DSA and Elgamal
(3) DSA (sign only)
(4) RSA (sign only)
Your selection?

Type 1 in the terminal and return:

RSA keys may be between 1024 and 4096 bits long.

What keysize do you want? (2048)

Return to take the default key long.

Please specify how long the key should be valid.

0 = key does not expire

<n> = key expires in n days

<n>w = key expires in n weeks

<n>m = key expires in n months

<n>y = key expires in n years

Key is valid for? (0)

Type return to take the default expire time, i.e. never expires:

Is this correct? (y/N)

Type y and continue with name and email:

You need a user ID to identify your key; the software constructs the user ID

from the Real Name, Comment and Email Address in this form:

"Heinrich Heine (Der Dichter) <heinrichh@duesseldorf.de>"

Real name: bozz

Name must be at least 5 characters long

Real name: Im Bozz

Email address:

Email address: bozz@goople.com

Comment: blabla

You selected this USER-ID:

"Im Bozz (blabla) <bozz@goople.com>"

Change (N)ame, (C)omment, (E)mail or (O)kay/(Q)uit?

type "O" and return to set 123 as password:

You need a Passphrase to protect your secret key.

gpg: gpg-agent is not available in this session

Enter passphrase:

123

gpg: gpg-agent is not available in this session

We need to generate a lot of random bytes. It is a good idea to perform

some other action (type on the keyboard, move the mouse, utilize the

disks) during the prime generation; this gives the random number

generator a better chance to gain enough entropy.

Not enough random bytes available. Please do some other work to give

the OS a chance to collect more entropy! (Need 236 more bytes)

Move your mouse, or type something with keybord in another terminal, and you will see the gpg process proceeds:

Not enough random bytes available. Please do some other work to give

the OS a chance to collect more entropy! (Need 104 more bytes)

..+++++

gpg: /home/tomy/.gnupg/trustdb.gpg: trustdb created

gpg: key 324B99FD marked as ultimately trusted

public and secret key created and signed.

gpg: checking the trustdb

gpg: 3 marginal(s) needed, 1 complete(s) needed, PGP trust model

gpg: depth: 0 valid: 1 signed: 0 trust: 0-, 0q, 0n, 0m, 0f, 1u

pub 2048R/324B99FD 2013-07-03

Key fingerprint = 5799 923C 4123 523E 0567 794F 2E52 B3F5 324B 99FD

uid Im Bozz (blabla) <bozz@goople.com>

sub 2048R/A57E911C 2013-07-03

Now the keys are generated.

Step4. Mr. Bozz try sign the contract again:

bozz@PC007x /home/bozz $ gpg --sign contract.txt

You need a passphrase to unlock the secret key for

user: "Im Bozz (blabla) <bozz@goople.com>"

2048-bit RSA key, ID 324B99FD, created 2013-07-03

gpg: gpg-agent is not available in this session

Enter passphrase:

Enter "123" as password, return. The signed contract file is created as "contract.txt.gpg"

bozz@PC007x /home/bozz $ ls
contract.txt contract.txt.gpg

Step5. Mr. Bozz send the "contract.txt.gpg" file to Tomy.

Step6. Tomy tries to verify file "contract.txt.gpg"

To verify the file is really from Mr. Bozz from GooPle, Tomy can use GPG utilities. We assume has GPG installed on his machine, and he has created his public and private key as described in Step.3

So Tomy types command below to verify the singnature of Mr. Bozz:

tomy@PC001x /home/tomy $ gpg --verify contract.txt.gpg

gpg: Signature made Tue 02 Jul 2013 03:00:40 PM CEST using RSA key ID B06521C9

gpg: Can't check signature: public key not found

The above information shows that gpg can not verify the signature, since it is missing the public key of Mr. Bozz public.

When Tomy tries to read or show the content of the contract "contract.txt.gpg", he will find that the file is actuall encrypted, and not readable.

So now Tomy needs to ask Mr. Bozz for his public key.

Step7. Mr. Bozz exports his public and send it to Tomy

Mr. Bozz exports his public key as file "bozz.asc" usingm command:

bozz@PC007x /home/bozz $ gpg -a --export > bozz.asc

A file "bozz.asc" is genereated with the content below:
===========
-----BEGIN PGP PUBLIC KEY BLOCK-----
Version: GnuPG v1.4.11 (GNU/Linux)

mQENBFHSzPABCADfOFyUJOpm+oMx0MgP6cZy1vMLkPpNNwMFTcdBxswx5t9w77XL
jPR1vw7GytLebmIYQ66NcJX3XETLHHcsSCEZF4ueAGbi/XSeBRWA/hSNGHRORkXa
Mvmq1uTq+aUbP/teOgfnEJCV1Rn5u1IP9W7GfJM7vNmfgBVwTj464zoePJjiF6Le
5Wwhkove2hURokltQJSJxFFfEEkEfrRqmAtzKGxxD2UXfAxp+/g1ULFl6J/1zDHN
Gdv9M6ppT9limlNCAwcRnas+bBvsyNF7sY+5CgX2rc6gusAcqXayqq1Ax5gnkE8V
14IJVyMbjeY+t84iWAi1eiva/Oq4b7W1dg/7ABEBAAG0Gm1vYmlsZWdhdGUgKG1n
KSA8bWdAbWcuZGU+iQE4BBMBAgAiBQJR0szwAhsDBgsJCAcDAgYVCAIJCgsEFgID
AQIeAQIXgAAKCRCYbbznsGUhycDmB/9+xPPIG2+Q5CqlmYqhgG3xsX0mQQW85w0r
LHJ3ix6Ah51u88OjvQ+57lkP+bE8EGOk6SGn/49OPM1DE5rAczDn9jdbfn8tVJ6k
pD1JKIPKC9yDhAmxGcRlzTPATJOYva4DMzNa7xFJTy4e97k2xgibESt0E6sh6EF5
ZtWJHyBAtdvCdZcLQy4bEAW6wOrGPv3T/tr/nW71hD96aGfiluRQ1Y3WfH15nc/u
CVEJlXmGP2NHHqYVjReMbPqEFfDlh+/tQ+0nvwYHc8p/Ixk6FbAUcbDXaNibZTZN
fS3F3CO7Oz/u5LM4CbOpTyqMSK6AlMU4kJECHdAAAcMyA4d2jM/guQENBFHSzPAB
CADAgZmDbU1PkJguIxq/j0fzJirGrE1k/U1rp4asQuKiU3WcVJ07h++yAwcvFz1 J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=49J7
-----END PGP PUBLIC KEY BLOCK-----

============

Mr. Bozz sends his public key, i.e. the file "bozz.asc" to Tomy

Step8. Tomy imports Mr. Bozz into gpg as trusted key:

tomy@PC001x /home/tomy $ gpg --import bozz.asc

Step9. Tomy tries to verify file "contract.txt.gpg" again:

tomy@PC001x /home/tomy $gpg --verify contract.txt.gpg

gpg: Signature made Tue 02 Jul 2013 03:00:40 PM CEST using RSA key ID B06521C9

gpg: Good signature from "Im Bozz (blabla) <bozz@goople.com>"

gpg: WARNING: This key is not certified with a trusted signature!

gpg: There is no indication that the signature belongs to the owner.

Primary key fingerprint: 850F 0E98 AF5E 4CD3 D9A9 0E1A 986D BCE7 B065 21C9

Step10. Tomy extracts the content from "contract.ext.gpg:

tomy@PC001x /home/tomy $ gpg --output my_contract.txt --decrypt contract.txt.gpg

gpg: Signature made Tue 02 Jul 2013 03:00:40 PM CEST using RSA key ID B06521C9

gpg: Good signature from "Im Bozz (blabla) <bozz@goople.com>"

gpg: WARNING: This key is not certified with a trusted signature!

gpg: There is no indication that the signature belongs to the owner.

Primary key fingerprint: 850F 0E98 AF5E 4CD3 D9A9 0E1A 986D BCE7 B065 21C9

A new file "my_contrct.txt" is created, Tomy can just show the file content with less command:

tomy@PC001x /home/tomy $ less my_contrct.txt

This is a contract between Mr. Tomy and GooPle ...

my_contrct.txt (END)

We are almost done here.
When we take another look at the verification information in step 9, we will see there is a warning says:

"WARNING: This key is not certified with a trusted signature! There is no indication that the signature belongs to the owner. "

This is because, the GPG utility of Tomy complains the public key of Mr. Bozz, is not certified by a known third party, a CA(Certificate Authority). The get away with the warning, Mr. Bozz need to have his public key file again signed by a thirty party, whom Tomy trust, i.e. Tomy had imported the 3 party's public key in his GPG utility.

We assume, Jerry is someone wo Tomy trust, so Tomy will import Jerry's public key by:

tomy@PC001x /home/tomy $ gpg --import jerry.asc

Step11. Mr. Bozz send his public key to Jerry

Step12. Jerry signs Mr. Bozz public key

Jerry imports Bozz's public key to his GPG utility:

jerry@PC001x /home/jerry $ gpg --import bozz.asc

Jerry sign Mr. Bozz' public key using :

jerry@PC001x /home/jerry $ gpg --sign-key B06521C9

Here B06521C9 is the id of Mr Bozz' key.

Jerry export Mr. Bozz's key as file:

jerry@PC001x /home/jerry $ gpg -a --export B06521C9 > bozz_signed.asc

The public key of Mr. Bozz is now signed by jerry, and stored in file "bozz_signed.asc"

Step13. Mr. Bozz signe the contract again.import the signed key "bozz_signed.asc"

Mr. Bozz imports the signed key sign the contract again, and send the contract to Tomy

Step14. Tomy verify the new contract file

tomy@PC001x /home/tomy $ gpg --verify samp.txt.gpg

gpg: Signature made Wed 03 Jul 2013 11:09:12 AM CEST using RSA key ID B06521C9

gpg: Good signature from "Im Bozz (blabla) <bozz@goople.com>"

Congratulations!

Initializing Log4j

2013-06-23T02:54:00.000-07:00

Document Version 1.0

You might have ever seen following warning running your program:

log4j:WARN No appenders could be found for logger (MyLogger).
log4j:WARN Please initialize the log4j system properly.

What does this mean?

It means Log4j is not initialized for your application.

Log4j could be initialized by using Log4j configurators. A more convenient way however is to provide a Log4j configuration file as JVM parameter, and let Log4j do the initialization. I will show you how does this work with a small program now:

Sample code:

package tryl4j;

import org.apache.log4j.Logger;

/**
* @author swang
*
*/
public class TryLog4jInit {
private static final Logger logger = Logger.getLogger("MyLogger");
public static void main(String[] ages) {
logger.info("Did you add a JVM param for initializing log4j?");
}

}

log4j.properties file

# Root logger option
log4j.rootLogger=INFO, stdout

# Direct log messages to stdout
log4j.appender.stdout=org.apache.log4j.ConsoleAppender
log4j.appender.stdout.Target=System.out
log4j.appender.stdout.layout=org.apache.log4j.PatternLayout
log4j.appender.stdout.layout.ConversionPattern=%d{yyyy-MM-dd HH:mm:ss} %-5p %c{1}:%L - %m%n

Run the above class with:

-Dlog4j.configuration="file:/tmp/log4j.properties"

And you will see the warning is not showing up and more, and you see the logged info in console:

2013-06-23 12:13:17 INFO MyLogger:16 - Did you add a JVM param for initializing log4j?

So the JVM parameter works!

JPQL vs. Criteria

2013-02-10T23:18:00.002-08:00

Document Version 1.0

JPA 2.0 introduced Criteria API for persistence entity manipulation. For those who has been working with JPQL, the question he might ask himself is: what does JPA Criteria has that JPQL NOT?

In the following I will try to compare JPQL and Criteria in the following 3 aspects:

Coding efforts
Usability
Performance

A simple "User.java" persistence entity will be used in test. To make the test simple, we will just test "SELECT" behavior of the JPQL and Criteria.

1. The persistence entity "User.java":

The "User.java" class has only 2 fields:

id, of type Long
name, of type String

2. Coding Effort: "SELECT" User

"SELECT" using JPQL:

...
Query q = em.createQuery("SELECT u From User u WHERE u.name=:name",
User.class);
q.setParameter("name", "hehe");

List<User> result = q.getResultList();
...

"SELECT" using Criteria API:

...

CriteriaQuery<User> query =cb.createQuery(User.class);
Root<User> aUser = query.from(User.class);
query.where(cb.equal(aUser.<String>get("name"), cb.parameter(String.class, "userName")));

TypedQuery<User> tq = em.createQuery(query);
tq.setParameter("userName", "hehe");

List<User> result = tq.getResultList();
...

Apparently the coding effort using JPQL is less than that when using Criteria.

3. Usability: "SELECT" User

For someone who is familar with SQL, he might feel very comfortable with JPQL syntax, however Criteria is not so straightforward.

For someone who use Criteria, he must also know what is SQL "from", "where".., so he is not going to have big problem with JPQL.

4. Performance

According to JPA documentation, "JPQL query is parsed for each call", where Criteria NOT, so criteria should perfoms better.

We will take data retrieve as example, to see if and how much is the performance diference actually.

Test run 1:

Test run 2:

The averave of 2 test runs:

The conclusion:

Deploy JAX-WS Web Service on Tomcat and JBoss

2013-02-08T00:05:00.001-08:00

Document Version 1.0

Keywords:

JAX-WS Web Service example, JAX-WS web service deployment Tomcat, JAX-WS web service deployment JBoss Glassfish, web service tutorial

Table of Contents

1. Create a Dynamic Web Project in Eclipse
2. Implement the "JaxWsSample" Web Service
3. Deploy the Web Service in JEE Container JBoss7
4.Test "HelloWorld" Web Service using SAOP UI
4.1 Download and install SOAP UI
4.2 Create a new SOAP UI project
4.3 Call "HelloWorld" Web Service
5. Deploy the Web Service in Tomcat 7
5.1 Download Tomcat Extras - "catalina-ws.jar"
5.2 Down JAX-WS RI lib
5.3 Configure JAX.WS Servlet and Listener
5.4 Add "sun-jaxws.xml"
5.5 Deploy "JaxWsSample.war" in Tomcat 7
5.6 Test "HelloWorld" Web Service using SAOP UI
5.7 Does the "HelloWorld" Web Service still work in JBoss 7 ?
6. To sum it up

With JAX-WS framework, developmet of Web Service is made very easier and straightworward. No boilerplate code like interfaces are needed any more, configuration is reduced to minmum... A developer can finally focus on the his work, i.e. the problem to solve.

However, dependend on if the underly container is a JEE container or Servlet container, the packaging, deplyoment, and even accessing of JAX-WS is also slightly different. In this post, we will show what are the differences exactly by implementing a simple web service, first deploly it in JBoss 7 application server, and than migrate to Servlet Container Tomcat 7.

1. Create a Dynamic Web Project in Eclipse

As development tool we will use "Eclipse IDE for Java EE Developers". We assume you have at least basic knowledge of using Eclipse IDE.

Following the steps to create a dynamic web project in Eclipse:

Select "Dynamic Web Project", click "Next"

Type "JaxWsSample" as Project Name. In the dropdown list of "Dynamic web module version", make sure "3.0" is selected. Click "Next".

Select checkbox "Genereate web.xml deployment descriptor", and click "Finish". Now you have created a web application project named "JaxWsSample" with following directory structure:

Click and open the file "web.xml", you will see the file is almost emtpy except the declaration of name space and schemas. For now will leave this file as it is, since no specific configuration is need to do here for deployment on JEE container.

2. Implement the "JaxWsSample" Web Service

We will create a simple "HelloWorld" web service which has only one operation "sayHello". As mentioned, for JAX-WS style web service, boilerplate code like interfaces are NOT needed any more. In our case, the impementation of the "HelloWorld" service contains only one class, i,e. "HelloWorld.java":

package sample.ws;

import javax.jws.WebMethod;
import javax.jws.WebParam;
import javax.jws.WebService;

@WebService()
public class HelloWorld {

@WebMethod //(operationName = "hello")
public String sayHello(@WebParam(name = "name") String name) {
return "Hello " + name + "!";
}
}

3. Deploy the Web Service in JEE Container JBoss 7

To deploy the "HelloWorld" service in a JBoss 7 container, there is no extra coding or configuration necessary.

What still need to be done is nothing else than packaging the "JaxWsSample" in a "war" archive, and put it in the "deployments" folder of a running JBoss 7 container.

Start JBoss 7 Application Server in standalone mode
Export "JaxWsSample" as "JaxWsSample.war"

The "Destination" should point to the "deployments" directory of the JBoss installation:

We assume the JBoss is up and running, then click "finish" Button in the above window will export the "JaxWsSinmple.war" to JBoss, and the "war" is deployed in JBoss.

Try opening following URL in your Web Browser "http://localhost:8080/JaxWsSample/HelloWorld?wsdl", you will be shown the "wsdl" file of the "HelloWorld" web service.

4. Test "HelloWorld" Web Service using SOAP UI

To test the "HelloWorld" Web Servcie, we need a Web Service Client. We will use SAOP UI for this.

4.1 Download and install SOAP UI

Download SAOP UI from www.soapui.org. After installation, start SOAP UI.

4.2 Create a new SAOP UI project

Create a Web Service (client) project using following WSDL location:
"http://localhost:8080/JaxWsSample/HelloWorld?wsd

Check the option "Create sample requests for all operations".

After click on the "OK" button, a "HelloWorld" web service client is created in SOAP UI. Open "Request 1".

4.3 Call "HelloWorld" Web Service

Replace the "?" with text "hehe" for the "name" tag of "Request 1". Click the green run button to execute the "sayHello" web service call.

The result of the web service call is show in the "response window":

5. Deploy the Web Service in Tomcat 7

In this section we will try to deploy the "HelloWorld" web service in a servlet container, i.e. Tomcat 7.

Different from JEE Conainer, Tomcat does NOT support JAX-WS out of box. To enable JAX-WS, followings have to be done:

include "catalina-ws.jar" in Tomcat lib
include "JAX-WS RI" libs
configure "jax-ws servlet" in web.xml
add "sun-jaxws-xml"

5.1 Download Tomcat Extras - "catalina-ws.jar"

Standard Tomcat delivery does NOT has libs supporting JAX-WS. We need to import "catalina-ws.jar" in Tomcat's "lib" directory. "catlina-ws.jar" belongs to Tomcat Extras, which could be downloaded here:

http://mirror.synyx.de/apache/tomcat/tomcat-7/v7.0.35/bin/extras/

5.2 Download JAX-WS RI libs

We need to include the JAX-WS runtime libs, i.e. JAX-WS RI (referencial implementation) in Tomcat. The lateste JAX-WS RI (2.2.7) could be downloaded here:

http://jax-ws.java.net/2.2.7/

Download and extract the arcive to you favorite location, for example to:

he@linux-7s1k:~/programms/jaxws-ri>
drwxr-xr-x 2 he users 4096 13. Aug 13:54 bin
-rw-r--r-- 1 he users 7405 13. Aug 13:54 build.xml
-rw-r--r-- 1 he users 38052 13. Aug 13:54 CDDL+GPLv2.txt
-rw-r--r-- 1 he users 1673 13. Aug 13:54 distributionREADME_WMforJava2.0.txt
drwxr-xr-x 6 he users 4096 13. Aug 13:54 docs
drwxr-xr-x 3 he users 4096 6. Jan 18:18 lib
-rw-r--r-- 1 he users 38052 13. Aug 13:54 LICENSE.txt
-rw-r--r-- 1 he users 1039 13. Aug 13:54 README
drwxr-xr-x 39 he users 4096 13. Aug 13:54 samples
-rw-r--r-- 1 he users 38607 13. Aug 13:54 ThirdPartyLicense.txt

The JAX-WS RI jars could be made available to Tomcat by including them in Tomcat's lib directory. Or you can include the jars in you Web Application's lib directory, in such case, the jars will be only available to the including Web Application.

We will take the second option and copy all jar files under

"/jaxws-ri/lib/"

to

"JaxWsSample" Eclipse project, and put the jars under

"/WebContent/WEB-INF/lib"

5.3 Configure jax-ws servlet

Change "web.xml" file by adding a "WSServletContextListener" and a "WSServlet":

<?xml version="1.0" encoding="UTF-8"?>
<web-app xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
   xmlns="http://java.sun.com/xml/ns/javaee"
   xmlns:web="http://java.sun.com/xml/ns/javaee/web-app_2_5.xsd"
   xsi:schemaLocation="http://java.sun.com/xml/ns/javaee
   http://java.sun.com/xml/ns/javaee/web-app_3_0.xsd"
  id="WebApp_ID" version="3.0">

<display-name>JaxWsSample</display-name>

<listener>
<listener-class>
com.sun.xml.ws.transport.http.servlet.WSServletContextListener
</listener-class>
</listener>

<servlet>
<servlet-name>hello</servlet-name>
<servlet-class>com.sun.xml.ws.transport.http.servlet.WSServlet</servlet-class>
<load-on-startup>1</load-on-startup>
</servlet>

<servlet-mapping>
<servlet-name>hello</servlet-name>
<url-pattern>/hello/*</url-pattern>
</servlet-mapping>
</web-app>

5.4 Add "sun-jaxws.xml"

Create a "sun-jaxws.xml" file in "WEN-INF" directory with following content:

<?xml version="1.0" encoding="UTF-8"?>
<endpoints xmlns="http://java.sun.com/xml/ns/jax-ws/ri/runtime" version="2.0">
<endpoint name="HelloWorld" implementation="sample.ws.HelloWorld" url-pattern="/hello"/>
</endpoints>

5.5 Deploy "JaxWsSample.war" in Tomcat 7

We assume the Tomcat 7 is up and running. Export the "JaxWsSample" project as "war" archive in Tomcat's "webapps" direcdory. The deployment will be successful ff you see following info in Tomcat console:

...
06.02.2013 19:21:49 org.apache.catalina.startup.HostConfig deployWAR
INFO: Deploying web application archive /home/he/apache-tomcat-7.0.29/webapps/JaxWsSample.war
06.02.2013 19:21:53 com.sun.xml.ws.transport.http.servlet.WSServletDelegate <init>
INFO: WSSERVLET14: Initialisierung von JAX-WS-Servlet
06.02.2013 19:21:53 com.sun.xml.ws.transport.http.servlet.WSServletContextListener contextInitialized
INFO: WSSERVLET12: Initialisierung von JAX-WS-Kontext-Listener
06.02.2013 19:21:53 com.sun.xml.ws.transport.http.servlet.WSServletContextListener contextInitialized
INFO: WSSERVLET12: Initialisierung von JAX-WS-Kontext-Listener

...

Try opening following URL in your Web Browser "http://localhost:8080/JaxWsSample/hello?wsdl", you will be shown the "wsdl" file of the "HelloWorld" web service.

Please NOTE, the URL to Web Service and WSDL is different as those in JBoss.
In Tomcat:
http://localhost:8080/JaxWsSample/hello?wsdl

In JBoss:
http://localhost:8080/JaxWsSample/HelloWorld?wsdl

5.6 Test "HelloWorld" web service using SAOP UI

Repeat section 4.2-4.3 to test "HelloWorld" web service, when create web service client in SOAP UI, be sure to use "http://localhost:8080/JaxWsSample/hello?wsdl" for WSDL location.

When you have followed all the steps without mistake till now, you would see the correct response of "sayHello".

5.7 Does the "HalloWorld" web service still work when deployed in JBoss 7

Now we have successed "migrated" a JAX-WS web service running on JBoss 7 to Tomcat 7. You might wounder, after all these changes, does the "HelloWorld" service still work in JBoss 7 ? We will find it out now.

Stop Tomcat 7, and hence stop "HelloWorld" in Tomcat.
Start JBoss 7 standardalone server
Copy "JaxWsSample.war" from Tomcat "webapps" directory to "deployments" directory of the JBoss installation.

You would see following info in JBoss 7 console:

...
21:31:33,602 INFO [org.jboss.wsf.stack.cxf.metadata.MetadataBuilder] (MSC service thread 1-2) Add Service
id=sample.ws.HelloWorld
address=http://localhost:8080/SampleWS/HelloWorld
implementor=sample.ws.HelloWorld
invoker=org.jboss.wsf.stack.cxf.JBossWSInvoker
serviceName={http://ws.sample/}HelloWorldService
portName={http://ws.sample/}HelloWorldPort
wsdlLocation=null
mtomEnabled=false
21:31:33,698 INFO [org.jboss.wsf.stack.cxf.metadata.MetadataBuilder] (MSC service thread 1-1) Add Service
id=sample.ws.HelloWorld
address=http://localhost:8080/JaxWsSample/HelloWorld
implementor=sample.ws.HelloWorld
invoker=org.jboss.wsf.stack.cxf.JBossWSInvoker
serviceName={http://ws.sample/}HelloWorldService
portName={http://ws.sample/}HelloWorldPort
wsdlLocation=null
mtomEnabled=false

The above info indicates that the deployment in JBoss till works!

Further more, you would find that the "HelloWorld" web service works too, and the service is even access able under 2 URLs:
http://localhost:8080/JaxWsSample/hello?wsdl
http://localhost:8080/JaxWsSample/HelloWorld?wsdl

6. To sum it up

Tutorial EJB3 Integration Test with Arquillian part4 - JBoss 5 managed container and TestNG

2012-12-29T22:02:00.002-08:00

Document Version 1.0

Keywords:
JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container, arquilllian TestNG

Part - 1 Part - 2 Part - 3 Part - 4

Table of Contents
1. Create a simple Maven project in Eclipse
2. Create project's pom.xml file
2.1 Replace the JUnit dedendency
2.2 Replaced the "arquillian-junit-container dedendency
3. Implement source code and test code
4. persistence.xml
5. java:/DefaultDS
6. arquillian.xml
7. Configuring TestNG "testng.xml"
8. Run test

I prefer TestNG to JUnit mainly for 2 reasons:

the various posibilities TestNG provides to configure and group tests.
the facilities of TestNG to test thread and concurrencies

Having shown Arquillian test with JUnit in the past, in this post we will see :

how to switch from JUnit to TestNG with Arquillian
JPA / Entity Bean Testing with Arquillian.

1. Create a simple Maven project in Eclipse

Follow the section 1 of "EJB3 Integration Test with Arquillian part1-JBoss 7 managed container" to create a new Eclipse project "testarq_jb5_managed_testng". Be sure in step2, to name the ArtifactId as "testarq_jb5_managed_testng".

2. Create project's pom.xml file

The "pom.xml" for TestNG is just slightly different from that of Turtorial part - 3:

2.1 Replace the JUnit dedendency:

<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<version>4.8.1</version>
<scope>test</scope>
</dependency>

with TestNG dependency:

<dependency>
<groupId>org.testng</groupId>
<artifactId>testng</artifactId>
<version>6.1.1</version>
<scope>test</scope>
</dependency>

2.2 Replaced the "arquillian-junit-container dedendency:

<dependency>
<groupId>org.jboss.arquillian.junit</groupId>
<artifactId>arquillian-junit-container</artifactId>
<version>${version.arquillian_core}</version>
<scope>test</scop>
</dependency>

with "arquillian-testng-container":

<dependency>
<groupId>org.jboss.arquillian.testng</groupId>
<artifactId>arquillian-testng-container</artifactId>
<version>${version.arquillian_core}</version>
</dependency>

3. Implement source code and test code

As code sample we will take a typical use case in JEE application:
A business layer/class uses a DAO (data access object) to access business data. Here business class and DAO class are both stateless session bean.

To implement and test the above use case, we need to create following classes in the "testarq_jb5_managed_testng" project:
2 for bussiness classes:
GrussGottEjbLocal.java
GrussGottEjb.java

2 for DAO classes:
UserDaoLocal.java
UserDao.java

The User Entity class:
and the test class:
User.java

And the test class:
TestGrussGott.java

Now the "testarq_jb5_managed_testng" project in Eclipse will looks like:

User.java

package test.ejb;
import java.io.Serializable;
import javax.persistence.Column;
import javax.persistence.Entity;
import javax.persistence.GeneratedValue;
import javax.persistence.Id;
import javax.persistence.Version;
@Entity
public class User implements Serializable {
private static final long serialVersionUID = 1L;
@Id
@GeneratedValue
private Long id;

@Column(nullable = false, unique = true)
private String name;

// private Long version;
@Version
private long version;

public Long getId() {
return id;
}
public void setId(Long id) {
this.id = id;
}
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
}

UserDaoLocal.java

package test.ejb;

import javax.ejb.Local;
@Local
public interface UserDaoLocal {
public void persist(User halloEntity);
public User findById(Long id);
public User findByName(String name);
}

UserDao.java

package test.ejb;
import javax.ejb.Stateless;
import javax.persistence.EntityManager;
import javax.persistence.PersistenceContext;
import javax.persistence.Query;
@Stateless
public class UserDao implements UserDaoLocal {
@PersistenceContext
EntityManager em;

@Override
public void persist(User halloEntity) {
em.persist(halloEntity);
}

@Override
public User findById(Long id) {
return em.find(User.class, id);
}

@Override
public User findByName(String name) {

String qfind = "select x from User x where x.name= :name";
Query findByName= em.createQuery(qfind);
findByName.setParameter("name", name);

return (User)findByNamet.getSingleResult();
}
}

GrussGottEjbLocal.java

package test.ejb;
import javax.ejb.Local;
@Local
public interface GrussGottEjbLocal {
public void newUser(User use);
public User findUserByName(String name);
}

GrussGottEjb.java

package test.ejb;
import javax.ejb.EJB;
import javax.ejb.Stateless;
@Stateless
public class GrussGottEjb implements GrussGottEjbLocal {
@EJB
UserDaoLocal userDao;
@Override
public void newUser(User user) {
userDao.persist(user);
}
@Override
public User findUserByName(String name) {
return userDao.findByName(name);
}
}

TestGrussGott.java

package test.ejb;

import javax.ejb.EJB;

import org.jboss.arquillian.container.test.api.Deployment;

import org.jboss.arquillian.testng.Arquillian;

import org.jboss.shrinkwrap.api.ShrinkWrap;

import org.jboss.shrinkwrap.api.spec.JavaArchive;

import org.testng.Assert;

import org.testng.annotations.Test;

/**

* @author he

// @RunWith(Arquillian.class)

public class TestGrussGott extends Arquillian {

@Deployment

public static JavaArchive createDeployment() {

JavaArchive jar = ShrinkWrap.create(JavaArchive.class, "hgrussEjb.jar");

jar.addPackage(User.class.getPackage());

jar.addAsResource("test-persistence.xml", "META-INF/persistence.xml");

// jar.addClass(GrussGottEjbLocal.class).addClass(GrussGottEjb.class);

// jar.addClass(UserDao.class).addClass(UserDaoLocal.class);

// jar.addAsManifestResource(EmptyAsset.INSTANCE, "beans.xml");

System.out.println(jar.toString(true));

return jar;

}

@EJB(mappedName = "test/GrussGottEjb/local")

GrussGottEjbLocal grusEjb;

// injecting em does not work for jb5 + arquillian, just use DAOs and

// business methods for inserting, deleting of entities

// @PersistenceContext

// EntityManager em;

@Test

public void sayHalloTxt() throws Exception {

User hehe = new User();

hehe.setName("hehe");

grusEjb.newUser(hehe);

User result = grusEjb.findUserByName("hehe");

Assert.assertEquals("hehe", result.getName());

System.out.println("#### found db entry text= " + result.getName()

+ " id=" + result.getId());

}

Explaination to "TestGrussGott.java"

Different from Arquillian JUnit test, "@RunWith(Arquillian.class)", is not supported by TestNG. For TestNG , test class should subclass "org.jboss.arquillian.testng.Arquillian".
Instead of adding class one by one to test archive, we add all classes in a package using " jar.addPackage(package_name)"
For JPA / Hibernate, test archive needs a "persistence.xml" file. This file should be added to test archive's "META-INF" folder. The line " jar.addAsResource("test-persistence.xml", "META-INF/persistence.xml");" says:

"add test-persistence.xml to test archive in META-INF folder, and rename the file to persistence.xml".

Use of EntityManager in test class

Whe you are familiar with persistence test with Springframework, you might remember, that you could inject an EntityManager in Spring test class, and use EntityManager to prepare test data. But unfortunately, for currently, Arquillian 1.0.3.Final + arquillian-jbossas-managed-5.1, does not support such a feature.

4. persistence.xml

As mentioned in section 3, Arquillian needs a "persistence.xml" for JPA / Hibernate to work. Such a "persistence.xml" is created only for test purpose, and is normally different from the one used in production environment.

We will create a "test-persistence.xml" file in "src/test/resources" folder. Arquillian takes care of renameing it to "persistence.xml" and packing this it in the deplyoment archive.

! You may wunder why we don't just name the original file as "persistence.xml"?

The answer is, we just want to avoid this file been taken mistakenly to be used for production environment.

The content of test-persistence.xml:

<?xml version="1.0" encoding="UTF-8"?>
<persistence version="2.0" xmlns="http://java.sun.com/xml/ns/persistence"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="
http://java.sun.com/xml/ns/persistence
http://java.sun.com/xml/ns/persistence/persistence_2_0.xsd">
<persistence-unit name="testPU">
<jta-data-source>java:/DefaultDS</jta-data-source>
<properties>
<property name="hibernate.hbm2ddl.auto" value="create-drop"/>
<property name="hibernate.show_sql" value="true"/>
</properties>
</persistence-unit>
</persistence>

5. java:/DefaultDS

The "persistence.xml" also contains data source configuration:

<jta-data-source>java:/DefaultDS</jta-data-source>

As data source will use the embedded hsqldb. The hsqldb is included in JBoss 5.1.0.
We just assume:

you have a "hsqldb-ds.xml" in "jbossHome/server/default/deploy",
the data source configuration works
the data source is bund to jndi "java:/defaultDS"

6. arquillian.xml

arquillian.xml should also be put in the folder "src/test/resources" :

<?xml version="1.0" encoding="UTF-8"?>
<arquillian xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns="http://jboss.org/schema/arquillian"
xsi:schemaLocation="http://jboss.org/schema/arquillian http://jboss.org/schema/arquillian/arquillian_1_0.xsd">

<container qualifier="jboss5" default="true">
<configuration>
<property name="jbossHome">/home/he/programms/jboss-5.1.0.GA</property>
<property name="profileName">default</property>
<property name="javaVmArguments">-Xmx512m -XX:MaxPermSize=256m</property>

<property name="jboss.server.log.threshold">INFO</property>
</configuration>
</container>
</arquillian>

7. Configuring TestNG "testng.xml"

We still need to configuration TestNG. TestNG provides the conveniece to group tests, and decide which tests are to be executed with changing the test classes.

We will create a "testng.xml" file in ""src/test/resources":

<!DOCTYPE suite SYSTEM "http://testng.org/testng-1.0.dtd" >
<suite name="Suite 1">
<test name="testNG test1">
<classes>
<class name="test.ejb.TestGrussGott" />
</classes>
</test>
</suite>

8. Run test

Now everything is set, we are ready to run the test.

Run Maven tests using following command

> mvn clean test -Parquillian-jbossas-managed_5_1

When you have followed all the steps till now, you would have seen test result like this in the console:

-------------------------------------------------------
T E S T S
-------------------------------------------------------
Running test.ejb.TestGrussGott
Configuring TestNG with: org.apache.maven.surefire.testng.conf.TestNGMapConfigurator@1cb25f1
30.12.2012 07:43:47 org.jboss.arquillian.container.impl.MapObject populate
WARNUNG: Configuration contain properties not supported by the backing object org.jboss.arquillian.container.jbossas.managed_5_1.JBossASConfiguration
Unused property entries: {jboss.server.log.threshold=INFO}
Supported property names: [shutdownTimeoutInSeconds, javaVmArguments, javaHome, partition, jbossHome, bindAddress, httpPort, startupTimeoutInSeconds, useRmiPortForAliveCheck, portBindingSet, profileName, rmiPort]
Starting server "default", with command (start timeout is 120 seconds ):
/usr/java/jdk1.6.0_32/bin/java -cp /home/he/programms/jboss-5.1.0.GA/bin/run.jar:/usr/java/jdk1.6.0_32/lib/tools.jar -Xmx512m -XX:MaxPermSize=256m -Djbosstest.udp.ip_ttl=0 -Djava.endorsed.dirs=/home/he/programms/jboss-5.1.0.GA/lib/endorsed -Djboss.server.log.threshold=DEBUG -Djava.endorsed.dirs=/home/he/programms/jboss-5.1.0.GA/lib/endorsed/ -Dxb.builder.useUnorderedSequence=true org.jboss.Main -c default -b localhost -g 13bea8d5080
Server started.
log4j:WARN No appenders could be found for logger (org.jboss.security.SecurityAssociation).
log4j:WARN Please initialize the log4j system properly.
hgrussEjb.jar:
/META-INF/
/META-INF/persistence.xml
/test/
/test/ejb/
/test/ejb/GrussGottEjbLocal.class
/test/ejb/UserDao.class
/test/ejb/GrussGottEjb.class
/test/ejb/TestGrussGott.class
/test/ejb/UserDaoLocal.class
/test/ejb/User.class
Shutting down server: default
shutdownTimeout will be=45
Server stopped.
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 162.169 sec

Results :

Tests run: 1, Failures: 0, Errors: 0, Skipped: 0

[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 3:16.269s
[INFO] Finished at: Sun Dec 30 07:46:25 CET 2012
[INFO] Final Memory: 28M/167M
[INFO] ------------------------------------------------------------------------

Part - 1 Part - 2 Part - 3 Part - 4

Tutorial EJB3 Integration Test with Arquillian part3 - JBoss 5 managed container

2012-12-18T12:04:00.001-08:00

Document Version 1.0

Keywords:
JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container, arquillian JUnit

Part - 1 Part - 2 Part - 3 Part - 4

Table of Contents

1. Create a simple Maven project in Eclipse
2. Create project's pom.xml file
3. Implement source code and test code
4. Download JBoss "jboss-5.1.0.GA"
5. arquillian.xml
6. Run the tests
6.1 Set system properties "PATH" and "JAVA_HOME"
6.2 Run tests6.2 Run tests
7. Finding the EJB with annotation
8. Run test again
9. Troubleshooting

In the part 1-2 of the "Tutorial EJB3 Integration Test with Arquillian" we successfully tested EJB with Arquillian, deployed on JBoss 7 managed and remote container.

But there are still cases, where upgrade to JBoss 7 is not an option, for whateve resons, therefore Arquillian test need to be done in older version of JBoss containers. Compare to the Arquillian test with JBoss 7, Arquillian test with older JBoss versionis are not so good documented.

In the part 3-5 of this serial I would try to to provide an "one stop" guide to setting up Arquillian with JBoss 5 container.

Arquillian part3 - JBoss 5 managed container

1. Create a simple Maven project in Eclipse

Follow the section 1 of "EJB3 Integration Test with Arquillian part1-JBoss 7 managed container" to create a new Eclipse project "testarq_jb5_managed". Be sure in step2, to name the ArtifactId as "testarq_jb5_managed".

2. Create project's pom.xml file

As you may remember, for Arquillian with JBoss 7, there is a dependency in pom.xml file called "arquillian-bom". We used this bom dependency to resolve all the Aquillian and JBoss artifacts. But unfortunately for JBoss 5, I could not find such a bom dependency, so all the dependencies in this pom.xml need to be configured "manually".

Create a "pom.xml" file in the "testarq_jb5_managed" project, which we created in section 1. Replace the content of "pom.xml" will the content below:

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">

<groupId>testarg</groupId>

<artifactId>testarq_jb5_1_managed</artifactId>

<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>

<version.arquillian_core>1.0.3.Final</version.arquillian_core>

<version.org.jboss.jbossas>5.1.0.GA</version.org.jboss.jbossas>

<version.org.jboss.jbossas_server-manager>1.0.3.GA</version.org.jboss.jbossas_server-manager>

<version.ejb3_api>3.1.0</version.ejb3_api>

</properties>

<!-- Use this repository you could find all jars for jb5. it compiles.

Test runs with error.. -->

<id>maven-us.nuxeo.org</id>

<url>https://maven-us.nuxeo.org/nexus/content/repositories/public</url>

</repository>

</repositories>

<build>

<artifactId>maven-compiler-plugin</artifactId>

</configuration>

</plugin>

<artifactId>maven-surefire-plugin</artifactId>

</plugin>

</plugins>

</build>

</dependencies>

<id>arquillian-jbossas-managed_5_1</id>

<groupId>junit</groupId>

<artifactId>junit</artifactId>

</dependency>

<groupId>org.jboss.arquillian.container</groupId>

<artifactId>arquillian-jbossas-managed-5.1</artifactId>

</dependency>

<groupId>org.jboss.jbossas</groupId>

<artifactId>jboss-server-manager</artifactId>

</dependency>

<groupId>org.jboss.jbossas</groupId>

<artifactId>jboss-as-client</artifactId>

<scope>provided</scope>

<groupId>javassist</groupId>

<artifactId>javassist</artifactId>

</exclusion>

</exclusions>

</dependency>

<groupId>javassist</groupId>

<artifactId>javassist</artifactId>

<scope>provided</scope>

</dependency>

<groupId>org.jboss.arquillian.junit</groupId>

<artifactId>arquillian-junit-container</artifactId>

<version>${version.arquillian_core}</version>

</dependency>

<groupId>org.jboss.arquillian.container</groupId>

<artifactId>arquillian-container-spi</artifactId>

<version>${version.arquillian_core}</version>

</dependency>

<groupId>trove</groupId>

<artifactId>trove</artifactId>

</dependency>

</dependencies>

</profile>

</profiles>

</project>

! Instead of using the default "maven central repository", we have to switch repository to "maven-us.nuxeo.org", as configured in the repositories section of the above "pom.xml" file.
Maven could not find some dependencies in "central maven repository" ( maybe they were removed, or not hosted there at all °_°).

3. Implement source code and test code

As code sample we will take a typical use case in JEE application:
A business layer/class uses a DAO (data access object) to access business data. Here business class and DAO class are both stateless session bean.

To implement and test the above use case, we need to create following classes in the "testarq_jb5_managed" project:
2 for bussiness classes:
GrussGottEjbLocal.java
GrussGottEjb.java

2 for DAO classes:
HalloDaoLocal.java
HalloDao.java

and the test class:
TestGrussGott.java

Now the "testarq_jb5_managed" project in Eclipse will looks like:

Replace the content of java files with following contents:

HalloDaoLocal.java
package test.ejb;

import javax.ejb.Local;
@Local
public interface HalloDaoLocal {
public String getText();
}

HalloDao.java

package test.ejb;

import javax.ejb.Stateless;

@Stateless

public class HalloDao implements HalloDaoLocal {

@Override

public String getText() {

return "!";

}

!
"HalloDao" is still not realy a DAO, it is not equipted with any persistence or database utilities.
Persistence with JPA / Hibernate will be shown in the part 4 of this serial. For now, we just want to simulate calling a session bean in an other session bean, therefore, no accessing database is really necessary.

GrussGottEjbLocal.java

package test.ejb;

import javax.ejb.Local;
@Local
public interface GrussGottEjbLocal {
public String sayHello(String name);
public String sayHelloTxt(String name);
}

GrussGottEjb.java
package test.ejb;

import javax.ejb.EJB;
import javax.ejb.Stateless;
@Stateless
public class GrussGottEjb implements GrussGottEjbLocal {
@EJB
HalloDaoLocal halloDao;
@Override
public String sayHello(String name) {
return "Hallo " + name;
}
@Override
public String sayHelloTxt(String name) {
String txt = halloDao.getText();
return "Hallo " + name + txt;
}
}

TestGrussGott.java

package test.ejb;

import javax.ejb.EJB;

import junit.framework.Assert;

import org.jboss.arquillian.container.test.api.Deployment;

import org.jboss.arquillian.junit.Arquillian;

import org.jboss.shrinkwrap.api.ShrinkWrap;

import org.jboss.shrinkwrap.api.asset.EmptyAsset;

import org.jboss.shrinkwrap.api.spec.JavaArchive;

import org.junit.Test;

import org.junit.runner.RunWith;

/**

* @author he

@RunWith(Arquillian.class)

public class TestGrussGott {

@Deployment

public static JavaArchive createDeployment() {

JavaArchive jar = ShrinkWrap.create(JavaArchive.class, "hgrussEjb.jar");

jar.addClass(GrussGottEjbLocal.class).addClass(GrussGottEjb.class);

jar.addClass(HalloDao.class).addClass(HalloDaoLocal.class);

//jar.addAsManifestResource(EmptyAsset.INSTANCE, "beans.xml");

System.out.println(jar.toString(true));

return jar;

}

@EJB

GrussGottEjbLocal grusEjb;

@Test

public void sayHallo() {

String result = grusEjb.sayHello("hehe");

Assert.assertEquals("Hallo hehe", result);

}

@Test

public void sayHalloTxt() {

String result = grusEjb.sayHelloTxt("hehe");

Assert.assertEquals("Hallo hehe!", result);

}

! In the "TestGrussGott.java" the line for "beans.xml" is commented out. This line is put here only to show, that unlike Arquillian test with JBoss 7 container, for Arquillian test with JBoss 5 container, the "beans.xml" is not needed.

4. Download JBoss "jboss-5.1.0.GA"

Dowload the JBoss server from www.jboss.org, and extract it to your favorite location.
On my machine , I extracted jboss to:

/home/he/jboss-5.1.0.GA

As usual, we need to set up PATH and JAVA_HOME system variables for jboss. When you have followed the part 1-2 of this turorial serial, where JBoss 7 is used, you mus have PATH and JAVA_HOME set to your jdk7 installation. This should be changed to a jdk6 location.
Following is the setting on my machine:

> export JAVA_HOME=/usr/java/jdk1.6.0_32/

> export PATH=/usr/java/jdk1.6.0_32/bin:$PATH

5. arquillian.xml

We still need to tell Arquillian where to find JBosss installation, and which JBoss instance is to be managed by Arquillian.

Create an "arquillian.xml" in the folder"src/test/resources" with the content below:

<?xml version="1.0" encoding="UTF-8"?>

<arquillian xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

xmlns="http://jboss.org/schema/arquillian"

xsi:schemaLocation="http://jboss.org/schema/arquillian http://jboss.org/schema/arquillian/arquillian_1_0.xsd">

<property name="jbossHome">path_2_your_jboss/jboss-5.1.0.GA</property>

<property name="profileName">default</property>

</configuration>

</container>

</arquillian>

6. Run the tests

We have learned how to run the Maven test in Eclipse. Now we will try runing the tests using command line.

6.1 Set system properties "PATH" and "JAVA_HOME"

Make sure you have correctly set "PATH" and "JAVA_HOME". In case of linux you could check this using following command:

> which java

/usr/java/jdk1.6.0_32/bin/java

> echo $JAVA_HOME

/usr/java/jdk1.6.0_32/

Command to set the properties:

export JAVA_HOME=/usr/java/jdk1.6.0_32/

export PATH=/usr/java/jdk1.6.0_32/bin:$PATH

6.2 Run tests

Run Maven tests using following command:

> mvn clean test -Parquillian-jbossas-managed_5_1

When you have followed all the steps till now, with no mistakes, the test results will be :

Test Failed!

To find out more details about the test failure, we will have a close look at the maven/surefile test report, this is the "test.ejb.TestGrussGott.txt" in /target/surefile-reports/:

...

-------------------------------------------------------------------------------

Test set: test.ejb.TestGrussGott

-------------------------------------------------------------------------------

Tests run: 2, Failures: 0, Errors: 2, Skipped: 0, Time elapsed: 124.531 sec <<< FAILURE!

sayHallo(test.ejb.TestGrussGott) Time elapsed: 1.442 sec <<< ERROR!

java.lang.NullPointerException

at test.ejb.TestGrussGott.sayHallo(TestGrussGott.java:48)

at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)

at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)

....

The Exception above says that in the "TestGrussGott,java", we have a problem on the line :

...
@Test

public void sayHallo() {

String result = grusEjb.sayHello("hehe");

Assert.assertEquals("Hallo hehe", result);

}
...

Here we have a NullPointerException for the object "grusEjb", i.e. "grusEjb" is null. This objected was injected in test class using @EJB annotation:

@EJB

GrussGottEjbLocal grusEjb;

We remember in the test cases using JBoss 7, we have always been doing like this, why this time failed?

Actually, our problem lies really in this place, in the @EJB annotation!
To put it short:

Arquillian can not find the "GrussGottEjb" using only @EJB, althrough the bean is deployed in JBoss.

7. Finding the EJB with annotation

When only @EJB is used, Arquillian tries to look up the "GrussGottEjb" session bean using following JNDI (could be seen in JBoss server.log):

....
FINE [org.jboss.arquillian.test.spi.TestEnricher]

(http-localhost%2F127.0.0.1-8080-2) Could not lookup @javax.ejb.EJB(beanName=, mappedName=, beanInterface=class java.lang.Object, description=, name=),

other Enrichers might, move on. Exception: No EJB found in JNDI,

tried the following names: java:global/test.ear/test/GrussGottEjbLocalBean,

java:global/test.ear/test/GrussGottEjbLocal,

java:global/test/GrussGottEjbLocal,

java:global/test/GrussGottEjbLocalBean,

java:global/test/GrussGottEjbLocal/no-interface,

test/GrussGottEjbLocalBean/local,

test/GrussGottEjbLocalBean/remote,

test/GrussGottEjbLocal/no-interface,

GrussGottEjbLocalBean/local,

GrussGottEjbLocalBean/remote,

GrussGottEjbLocal/no-interface,

ejblocal:test.ejb.GrussGottEjbLocal,

test.ejb.GrussGottEjbLocal,
....

But, when the "hgrussEjb.jar" is deployed by Arquillian in JBoss, it is bund to the following JNDI as shown in JBoss "server.log":

...

jndi:test/GrussGottEjb/local-test.ejb.GrussGottEjbLocal

jndi:test/GrussGottEjb/local

jndi:test/GrussGottEjb/remote
...

Now problem is clear: using @EJB only is not enough!
The solution is actually easy. In the "TestGrussGott,java", just change the annotation from:

@EJB

GrussGottEjbLocal grusEjb;

to :

@EJB(mappedName="test/GrussGottEjb/local")

GrussGottEjbLocal grusEjb;

8. Run test again

Now run the test again:

> mvn clean test -Parquillian-jbossas-managed_5_1

And you would see in the console:

9. Troubleshooting

9.1 Exception

....
Failed to collect dependencies for....
...
Could not transfer artifact org.jboss.aop:jboss-aop-aspects:pom:2.1.1.GA from/to repository.jboss.org (http://repository.jboss.org/maven2):
Access denied to: http://repository.jboss.org/maven2/org/jboss/aop/jboss-aop-aspects/2.1.1.GA/jboss-aop-aspects-2.1.1.GA.pom,

...

This indicates that dependencies could not be downloaded from Maven central repository. Check your pom.xml file in section 2 of this turorial, make sure you have configured /customized Maven repository (for example "maven-us.nuxeo.org"" ).

9.2 When you observer error information like:

Error in Console:

Tests in error:

test.ejb.TestGrussGott: Could not start remote container

Error in "test.ejb.TestGrussGott.tx":

org.jboss.arquillian.container.spi.client.container.LifecycleException: Could not start remote container

...

Caused by: java.io.IOException: Server failed to start; see logs. exit code: 0
...

And Error in JBoss "server.log"

Failed to get property value for bean: org.jboss.web.tomcat.service.management.ConnectorBean, property: bytesSent
java.lang.IllegalStateException: java.lang.RuntimeException: Error creating MBeanProxy: jboss.web:name=http-localhost-8080,type=GlobalRequestProcessor

...

Caused by: java.lang.RuntimeException: Error creating MBeanProxy: jboss.web:name=http-localhost-8080,type=GlobalRequestProcessor

The reason for the above errors is very likely running JBoss with the wrong Java version. So please make sure to run JBoss with jdk6, anot NOT jdk7. Check PATH and JAVA_HOME of your system again.

Part - 1 Part - 2 Part - 3 Part - 4

Tutorial EJB3 Integration Test with Arquillian part2 - JBoss 7 remote container

2012-12-17T11:12:00.003-08:00

Document Version 1.0

Keywords:
JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container

Part - 1 Part - 2 Part - 3 Part - 4

Table of Contents
1. Create a simple Maven project in Eclipse
2. Create project's pom.xml file
3. Implement source code and test code
4. Start JBossjboss-as-7.1.1.Final
5. arquillian.xml
6. Run Arquillian test
7. Troubleshooting

When you have followed the part1 of this serial "Tutorial Integration test with Arquillian" with success, your system would have been correctly set up to run further Arquillian test, - with some samll adaptions depending upon the container you are planning to use.

In this part2 of the serial we will continue with part1, to test stateless session bean in a JBoss 7 remote container. Instead of reusing the "testarq_jb7_managed" Eclipse project we created in part1, we will create a new Eclipse project "testarq_jb7_remote", reuse the EJB classes and the test class.

1. Create a simple Maven project in Eclipse

Follow the section 1 of "EJB3 Integration Test with Arquillian part1-JBoss 7 managed container" to create a new Eclipse project ""testarq_jb7_remote". Be sure in step2, to name the ArtifactId as ""testarq_jb7_remote".

2. Create project's pom.xml file.

Replace the content of pom.xml file with the following content:

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">

<groupId>testarg</groupId>

<artifactId>testarq_jb7_remote</artifactId>

<description>test arrquillian</description>

<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>

</properties>

<groupId>org.jboss.arquillian</groupId>

<artifactId>arquillian-bom</artifactId>

<version>1.0.3.Final</version>

<scope>import</scope>

</dependency>

</dependencies>

</dependencyManagement>

<build>

<artifactId>maven-compiler-plugin</artifactId>

</configuration>

</plugin>

<artifactId>maven-surefire-plugin</artifactId>

</plugin>

</plugins>

</build>

<groupId>junit</groupId>

<artifactId>junit</artifactId>

</dependency>

<groupId>org.jboss.arquillian.junit</groupId>

<artifactId>arquillian-junit-container</artifactId>

</dependency>

<groupId>org.jboss.spec</groupId>

<artifactId>jboss-javaee-6.0</artifactId>

<version>1.0.0.Final</version>

<scope>provided</scope>

</dependency>

</dependencies>

<id>arquillian-jbossas-remote_7</id>

<groupId>org.jboss.as</groupId>

<artifactId>jboss-as-arquillian-container-remote</artifactId>

<version>7.1.1.Final</version>

</dependency>

</dependencies>

</profile>

</profiles>

</project>

3. Implement source code and test code

Follow the section 3 of "EJB3 Integration Test with Arquillian part1-JBoss 7 managed container" to create the 2 classes for Hallo EJB and test class:

HalloEjbLocal.java

HalloEjb.java

TestHalloEJB.java

4. Start JBossjboss-as-7.1.1.Final

Again here make sure you have system property PATH including "bin" folder of you jdk7 location, and JAVA_HOME pointing to jkd7 installation location.

Unlike doing Arquillian test with managed container, where Arquillian takes care of start and stop of remote server. When running with Arquillian with remote server, before starting test, the remote server must be already started and up.

Start JBoss7 in standalone mode:

he@linux-7s1k:~/jboss-as-7.1.1.Final/bin> ./standalone.sh

======================================================================

JBoss Bootstrap Environment

JBOSS_HOME: /home/he/jboss-as-7.1.1.Final

JAVA: /var/jdk7/bin/java

JAVA_OPTS: -server -XX:+TieredCompilation -Xms64m -Xmx512m -XX:MaxPermSize=256m -Djava.net.preferIPv4Stack=true -Dorg.jboss.resolver.warning=true -Dsun.rmi.dgc.client.gcInterval=3600000 -Dsun.rmi.dgc.server.gcInterval=3600000 -Djboss.modules.system.pkgs=org.jboss.byteman -Djava.awt.headless=true -Djboss.server.default.config=standalone.xml

======================================================================

5. arquillian.xml

In case of remote JBoss container, Arquillian needs 2 extra parameters: managementAddress managementPort.

Arquillian use default "127.0.0.1:9999", if no configuration is found.

If your server is running on different management host/port, then the 2 parameters should be configured in arquillian.xml

6. Run Arquillian test

Create a Ecliipse Run configuration for Maven as shown in the following image:

An Click Run button.

When you have correctly followed the section 1-6, you would see in you console a screen likes:

When you take a look at the JBoss 7 log file, you would see the "halloEjb.jar" is deployed at the beginning of the test, and undeployed from JBoss when the test is finished:

...
20:25:04,190 INFO [org.jboss.as.server.deployment] (MSC service thread 1-3) JBAS015876: Starting deployment of "halloEjb.jar"
...
20:25:04,338 INFO [org.jboss.as.ejb3.deployment.processors.EjbJndiBindingsDeploymentUnitProcessor] (MSC service thread 1-4) JNDI bindings for session bean named HalloEjb in deployment unit deployment "halloEjb.jar" are as follows:
java:global/halloEjb/HalloEjb!test.ejb.HalloEjbLocal
java:app/halloEjb/HalloEjb!test.ejb.HalloEjbLocal
java:module/HalloEjb!test.ejb.HalloEjbLocal
java:global/halloEjb/HalloEjb
java:app/halloEjb/HalloEjb
java:module/HalloEjb
...
20:25:05,142 INFO [org.jboss.as.server] (management-handler-thread - 2) JBAS018559: Deployed "halloEjb.jar"
20:25:06,750 INFO [org.jboss.weld.deployer] (MSC service thread 1-1) JBAS016009: Stopping weld service for deployment halloEjb.jar
....
20:25:07,152 INFO [org.jboss.as.server] (management-handler-thread - 1) JBAS018558: Undeployed "arquillian-service"
...

7. Troubleshooting

7.1 When you see test failure like:

...
Tests in error:
test.ejb.TestHalloEJB: Could not create a new instance of class org.jboss.arquillian.test.impl.EventTestRunnerAdaptor
....
Caused by: java.lang.NoClassDefFoundError: org/jboss/arquillian/test/spi/enricher/resource/ResourceProvider
...
Caused by: java.lang.RuntimeException: Could not create a new instance of class org.jboss.arquillian.core.impl.ManagerImpl see cause.
...

This is most possibly because the "aqruillian.bom" version is not compatible with the "jboss-as-arquillian-container-remote" version. Please make sure you used the version shown in the pom.xml of this tutorial.

7.2 When you see test failure like:

...
org.jboss.shrinkwrap.api.exporter.ArchiveExportException: Failed to write asset to output: /junit/extensions/TestDecorator.class
...
Caused by: java.io.IOException: Pipe closed
...
Possible deadlock scenario: Got exception on closing the ZIP out stream: Pipe closed
...
Failed to write asset to output: /test/ejb/TestHalloEJB.class
...

Tests in error:
test.ejb.TestHalloEJB: Could not deploy to container: JBAS012144: Could not connect to remote://127.0.0.1:9999. The connection timed out
...

This normally indicates that the JBoss server is not running, or it runs with different "managementAddress" or "managementPort" than default. Please check your configuration against the section 5 of this tutorial.

Tutorial EJB3 Integration Test with Arquillian part3 - JBoss 5 managed container

Part - 1 Part - 2 Part - 3 Part - 4

Tutorial EJB3 Integration Test with Arquillian part1 - JBoss 7 managed container

2012-12-15T22:24:00.005-08:00

Document Version 1.0

Keywords:
JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container

Part - 1 Part - 2 Part - 3 Part - 4

Table of Contents
1. Create a simple Maven project in Eclipse
2. Create project's pom.xml file
3. Implement source code and test code
3.1 HalloEjbLocal.java
3.2 HalloEjb.java
3.3 TestHalloEJB.java
4. Download JBossjboss-as-7.1.1.Final
5. Run the test
6. Configuration Arquillian to find the JBoss Server

W ith Arquillian becoming more and more known to the JEE world, more and more people may have tried to evaluate it, to find if it could be the EJB testing tool for the next project. But it often happens, after hours of trying, we still have no luck making even a Hello World test to work. We might have problem finding all the necessary dependencies, we might miss correct configuration files, we might have written the test classes slightly different from what was shown in the official guids...

All these could happen, if you are not using the "standard combinations" shown in the official Arquillian documents. For example, currently the combination of "arquillian 1.0.0.3Final + jboss7 embedded + maven + JUnit" is quite well documented. But, if you have jboss7 managed or remote container, even jboss5 container, if your prefere TestNG over JUnit, or you are using ANT not maven, in such cases, the way to Arquillin Hello world are often not so straightforward, and you might think of giving it up.

In the upcomming "EJB3 Integration Test with Arquillian" serials, I will try to provide several step for step running samples, to show how to setup some standard and non-stand combinations of Arquillian usage. This includes:

Tutorial EJB3 Integration Test with Arquillian part1 - JBoss 7 managed container
Tutorial EJB3 Integration Test with Arquillian part2 - JBoss 7 remote container
Tutorial EJB3 Integration Test with Arquillian part3 - JBoss 5 managed container
Tutorial EJB3 Integration Test with Arquillian part4 - JBoss 5 managed container and TestNG (Testing of JPA/Entity Bean using Arquillian is also included in this part)
Tutorial EJB3 Integration Test with Arquillian part5 - JBoss 5 managed container, TestNG and ANT(+ Ivy)

To follow this "Tutorial EJB3 Integration Test with Arquillian" serial, you should have at least of basic knowledges of :
Using Eclipse and Maven
JEE/EJB3 programming model
JUnit

Now let's start with the part1

Tutorial EJB3 Integration Test with Arquillian part1 - JBoss 7 managed container

In this section we will show an integration test of EJB3 stateless session bean using Arquillian with managed JBoss 7 container.

1. Create a simple Maven project in Eclipse

We assume you have Eclipse JEE IDE with Maven plugin. Follow the steps below to create a simple Maven project in Eclipse.

step1. Eclipse -> new Project -> Maven Project:

select check-box "Create a simple project" and "Use default Workspace location". Click "Next".

step2. in the "Maven new Project" view, provide "testarq" as Group Id and "testarq_jb7_managed" as Artifact Id.

Give "0.0.1" as Version, leave default "jar" as Packaging (type).

Click "finish".

A new Maven project called "testarq_jb7_managed" is created, and show in Eclipse:

2. Create project's pom.xml file

Replace the content of pom.xml file with the following content:

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"

xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">

<groupId>testarg</groupId>

<artifactId>testarq_jb7_managed</artifactId>

<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>

</properties>

<groupId>org.jboss.arquillian</groupId>

<artifactId>arquillian-bom</artifactId>

<version>1.0.3.Final</version>

<scope>import</scope>

</dependency>

</dependencies>

</dependencyManagement>

<build>

<artifactId>maven-compiler-plugin</artifactId>

</configuration>

</plugin>

<artifactId>maven-surefire-plugin</artifactId>

</plugin>

</plugins>

</build>

<groupId>junit</groupId>

<artifactId>junit</artifactId>

</dependency>

<groupId>org.jboss.arquillian.junit</groupId>

<artifactId>arquillian-junit-container</artifactId>

</dependency>

<groupId>org.jboss.spec</groupId>

<artifactId>jboss-javaee-6.0</artifactId>

<version>1.0.0.Final</version>

<scope>provided</scope>

</dependency>

</dependencies>

<id>arquillian-jbossas-managed</id>

<groupId>org.jboss.arquillian.junit</groupId>

<artifactId>arquillian-junit-container</artifactId>

</dependency>

<groupId>org.jboss.spec</groupId>

<artifactId>jboss-javaee-6.0</artifactId>

<version>1.0.0.Final</version>

<scope>provided</scope>

</dependency>

<groupId>org.jboss.as</groupId>

<artifactId>jboss-as-arquillian-container-managed</artifactId>

<version>7.1.1.Final</version>

</dependency>

<groupId>org.jboss.arquillian.protocol</groupId>

<artifactId>arquillian-protocol-servlet</artifactId>

</dependency>

</dependencies>

</profile>

</profiles>

</project>

For the example to work, please make sure you have arquillian-bom version "1.0.0.Final" and "jboss-as-arquillian-container-managed" with version "7.1.1.Final".

Correct arquillian-bom version helps to resolve the needed dependencies. When you have problem like "ClassNotFoundException" or ".. one of its dependencies could not be resolved: Failed to read artifact ..."

You have probabily problem with arquillian-bom version (Can also be that the dependencies are not in the (default) maven central repository, then you need to configure maven to use another repository, we will talk about this issue in detail in another post). Try to use a correct arquillian-bom, instead of adding dependencies manually.

3. Implement source code and test code

Now we ready to write the EJB and test codes.

We are going to write a HalloEJB stateless session bean. This bean has one method "sayHello(txt)", which return "Hallo xxx" text.

The bean itself consistes of 2 java classes. i.e. HalloEjb.java and HalloEjbLocal.java. We also need a test class TestHelloEJB.java

3.1 HalloEjbLocal.java

package test.ejb;

import javax.ejb.Local;

@Local

public interface HalloEjbLocal {

public String sayHello(String name);

}

3.2 HalloEjb.java

package test.ejb;

import javax.ejb.Stateless;

@Stateless
public class HalloEjb implements HalloEjbLocal {
public String sayHello(String name) {
return "Hallo " + name;
}
}

3.3 TestHalloEJB.java

package test.ejb;

import javax.ejb.EJB;
import junit.framework.Assert;
import org.jboss.arquillian.container.test.api.Deployment;
import org.jboss.arquillian.junit.Arquillian;
import org.jboss.shrinkwrap.api.ShrinkWrap;
import org.jboss.shrinkwrap.api.asset.EmptyAsset;
import org.jboss.shrinkwrap.api.spec.JavaArchive;
import org.junit.Test;
import org.junit.runner.RunWith;

@RunWith(Arquillian.class)
public class TestHalloEJB {
@Deployment
public static JavaArchive createDeployment(){
JavaArchive jar = ShrinkWrap.create(JavaArchive.class, "halloEjb.jar");
jar.addClass(HalloEjbLocal.class).addClass(HalloEjb.class);
jar.addAsManifestResource(EmptyAsset.INSTANCE, "beans.xml");
//jar.addAsManifestResource("ejb-jar.xml", "ejb-jar.xml");
System.out.println(jar.toString(true));

return jar;
}

@EJB
HalloEjbLocal halloEjb;
@Test
public void sayHallo(){
String result = halloEjb.sayHello("hehe");
Assert.assertEquals("Hallo hehe", result);
}
}

The @Deployment annotation in the above tells Arquillian at runntimeto assemble a "halloEjb.jar" and deplyoment this to the assigned container.

The line "jar.addAsManifestResource(EmptyAsset.INSTANCE, "beans.xml");" is very important here. Other wise, the deployment will fail.(We will see in subsequent post, that the "bean.xml" is sometimes not needed, if the used dependency version is diffeeren than Arquillian 1.0.0.Final + jboss-as-arquillian-container-managed 7.1.1.Final ).

The way EJBs are referenced here conforms with EJB3 annotation style @EJB, just like EJB references in normal EJB3 conponents. No JNDI lookup code is necessary in test code, and that is the magic of Arquillian!

When you have followed the steps till now, you would have an Eclipse project as shown in the following image:

As mentioned, we will run the EJB test in a "managed jboss7 conteiner". This means, at run time Arquillian will start an JBoss7 container, deploy the to be tested code in the JBoss, execute the test, and then shutdown the JBoss7 instance.

So we still need a JBoss7 server.

4. Download JBossjboss-as-7.1.1.Final

Dowload the JBoss server from www.jboss.org, and extract it to your favorite location.
On my machine , I extracted jboss to:
/home/he/jboss-as-7.1.1.Final

In order to run jboss you still need to have add "bin" directory of jdk7 to the PATH of your system, and set JAVA_HOME to the jdk7 installation location.

Now we are "ready" to run the "TestHalloEJB" test

5. Run the test

Create an Eclipse Run configuration for maven projects as shown below:

Make sure you have used the "arquillian-jbossas-managed" profile, and Click run.

You could also run test on commandline using following command:

> mvn clean test -Parquillian-jbossas-managed

When you have correctly followed the step 1-5 till now you would see following screen in your console:

We have test error!

When you have a close look at the error text, you will see, Arquillian complains aobut missing "jbossHome". So, this is the problem.

We need to tell Arquillian where is the JBosss 7 server.

6. Configuration Arquillian to find the JBoss Server

Configuring Arquillian to find JBoss could be down in 2 ways. You could either set system property "JBOSS_HOME", or use "arquillian.xml". Using "arquillian.xml" file will provide us more felxibility, so we will take this option.

Create an "arquillian.xml" in "/src/test/resources/" folder with following content:

<arquillian xmlns="http://jboss.org/schema/arquillian"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://jboss.org/schema/arquillian
http://jboss.org/schema/arquillian/arquillian_1_0.xsd">
<container qualifier="jbossas-7-managed" default="true">
<configuration>
<property name="jbossHome">/home/he/jboss-as-7.1.1.Final</property>
</configuration>
</container>
</arquillian>

Please remember to replace "/home/he/jboss-as-7.1.1.Final" with the real JBoss7 location on your system

Now try to runt the test again. When you have not made any mistake following the above steps, you will see that the test are successful in your console:

-------------------------------------------------------

T E S T S

-------------------------------------------------------

Forking command line: /bin/sh -c cd /home/he/workspace_arqtest/testarq_jb7_managed && /usr/java/jdk1.6.0_32/jre/bin/java -jar /home/he/workspace_arqtest/testarq_jb7_managed/target/surefire/surefirebooter7540012828763289102.jar /home/he/workspace_arqtest/testarq_jb7_managed/target/surefire/surefire6295545167876887576tmp /home/he/workspace_arqtest/testarq_jb7_managed/target/surefire/surefire_07068526860728801090tmp

Running test.ejb.TestHalloEJB

log4j:WARN No appenders could be found for logger (org.jboss.logging).

log4j:WARN Please initialize the log4j system properly.

16.12.2012 21:11:08 org.jboss.as.arquillian.container.managed.ManagedDeployableContainer startInternal

INFO: Starting container with: [/var/jdk7/bin/java, -Xmx512m, -XX:MaxPermSize=128m, -ea, -Djboss.home.dir=/home/he/jboss-as-7.1.1.Final, -Dorg.jboss.boot.log.file=/home/he/jboss-as-7.1.1.Final/standalone/log/boot.log, -Dlogging.configuration=file:/home/he/jboss-as-7.1.1.Final/standalone/configuration/logging.properties, -Djboss.modules.dir=/home/he/jboss-as-7.1.1.Final/modules, -Djboss.bundles.dir=/home/he/jboss-as-7.1.1.Final/bundles, -jar, /home/he/jboss-as-7.1.1.Final/jboss-modules.jar, -mp, /home/he/jboss-as-7.1.1.Final/modules, -jaxpmodule, javax.xml.jaxp-provider, org.jboss.as.standalone, -server-config, standalone.xml]

21:11:17,820 INFO [org.jboss.modules] JBoss Modules version 1.1.1.GA

....

21:12:03,669 INFO [org.jboss.as] (Controller Boot Thread) JBAS015874: JBoss AS 7.1.1.Final "Brontes" started in 50871ms - Started 187 of 264 services (76 services are passive or on-demand)
halloEjb.jar:
/META-INF/
/META-INF/beans.xml
/test/
/test/ejb/
/test/ejb/HalloEjb.class
/test/ejb/HalloEjbLocal.class
...

21:12:12,472 INFO [org.jboss.as.server.deployment] (MSC service thread 1-4) JBAS015876: Starting deployment of "halloEjb.jar"

21:12:12,643 INFO [org.jboss.weld.deployer] (MSC service thread 1-4) JBAS016002: Processing weld deployment halloEjb.jar

21:12:12,689 INFO [org.jboss.as.ejb3.deployment.processors.EjbJndiBindingsDeploymentUnitProcessor] (MSC service thread 1-4) JNDI bindings for session bean named HalloEjb in deployment unit deployment "halloEjb.jar" are as follows:

java:global/halloEjb/HalloEjb!test.ejb.HalloEjbLocal

java:app/halloEjb/HalloEjb!test.ejb.HalloEjbLocal

java:module/HalloEjb!test.ejb.HalloEjbLocal

java:global/halloEjb/HalloEjb

java:app/halloEjb/HalloEjb

java:module/HalloEjb

21:12:12,882 INFO [org.jboss.weld.deployer] (MSC service thread 1-1) JBAS016005: Starting Services for CDI deployment: halloEjb.jar

....

Results :
Tests run: 1, Failures: 0, Errors: 0, Skipped: 0
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 1:43.141s
[INFO] Finished at: Sun Dec 16 21:12:17 CET 2012
[INFO] Final Memory: 23M/186M
[INFO] ------------------------------------------------------------------------

Congratulations!
You have successfuly tested a stateless session bean using Arquillian with managed JBoss 7.1.1.GA container

Tutorial EJB3 Integration Test with Arquillian part2 - JBoss 7 remote container

Part - 1 Part - 2 Part - 3 Part - 4

Hibernate MultipleBagFetchException simultaneously fetch multiple bags

2012-10-25T04:33:00.000-07:00

Document Version 1.0

JBoss Dynamic Migration of Distributed HA Singletons among Nodes

2012-10-18T11:47:00.002-07:00

Document Version 1.0

Keywords:
JBoss HA Service, HA Singleton, JBoss Cluster, JBoss load balancing, load migration, load distribution

A JBoss HA singleton is a cluster wide singleton, which runs only on one node of a JBosss cluster. When the node on which the singeton runs, fails, another node (the master node) will automatically start the singelton service on it. When there are several different HA singleton services, ALL the services will be activated on the master node. When the master node fails, another node take over the role of of master node, and start ALL the singleton services on this node. This is the default HA singleton behavior of JBoss.

However there might be some problems with the above solution.
If the singleton services, or some of the singleton services are resource intensive, we are going to have performance problem, if they all run simutaniously on one node, while other nodes are relatively idle.

We will now introduce a solution, which dynamically distribute HA singleton services on preferred nodes on cluster startup. In case of server failures, services will be migrated and distributed to other nodes, by checking the number of still alive nodes, and by considering the preferred nodes.

With an example, we will show how 3 HA singleton services could be dynmically migrated among a 3-nodes JBoss cluster.

Rules for distributing services on nodes are:

if alive node count 3 -> TestHASvcA on node[0]
TestHASvcB on node[1]
TestHASvcC on node[2]

if alive node count 2 -> TestHASvcA on node[0]
TestHASvcB on node[0]
TestHASvcC on node[1]

if alive node count 1 -> TestHASvcA on node[0]
TestHASvcB on node[0]
TestHASvcC on node[0]

Explaination to node sequences:
For a JBoss cluster, all nodes share the same sorted list of all alive nodes. This list is sorted by the time, when a node joins the cluster. For a 3-nodes cluster, node1, ndoe2, and node3, if nodes are started one after another, then the order list will be :
node1 --> [0]
node2 --> [1]
node3 --> [2]

In case node1 fails, node2 and node3 will update the list to:
node2 --> [0]
node3 --> [1]

When node1 joins the cluster again (restarted), the shared list of the 3 nodes is changed to:
node2 --> [0]
node3 --> [1]
node1 --> [2]

Expected HA service migration behavior of our example :

step1. start node1 TestHASvcA on node1 TestHASvcB on node1 TestHASvcC on node1	step2. start node2 TestHASvcA on node1 TestHASvcB on node1 TestHASvcC migrates to node2	step3. start node3 TestHASvcA still on node1 TestHASvcB migrates to node2 TestHASvcC migrates to node3

step6. node1 recovers TestHASvcA still on node3 TestHASvcB still on node3 TestHASvcC migrates to node1	step5. node2 fails TestHASvcA migrates to node3 TestHASvcB migrates to node3 TestHASvcB still on node2 TestHASvcC still on node3	step4. node1 fails TestHASvcA migrates to node2 TestHASvcB still on node2 TestHASvcC still on node3

step7. node2 recovers TestHASvcA still on node3 TestHASvcB migrate to node1 TestHASvcC migrates to node2

Configure and run a 3-nodes JBoss cluster

Please refer to the post HA Singleton, Cluster Wide Singleton as MBean in JBoss 5, part 1/3, and create a JBoss cluster with 3 nodes:

node1
node2
node3

Create 3 MBean TestHASvcA, TestHASvcB, TestHASvcC.

Use your favorite IDE to creat a simple Java project "HAServiceDynamicLoadMigration".
Please refer to the post HA Singleton, Cluster Wide Singleton as MBean in JBoss 5, part 2/3, and create 3 JBoss MBeans TestHASvcA, TestHASvcB and TestHASvcC.

TestHASvcA.java

package test.ha;
import org.jboss.system.ServiceMBeanSupport;
/**
* Make sure to use the naming standard for MBeans (If the
* source class is named Serious, then the interface must be named
* SeriousMBean).
*
* @author ws
*/
public class TestHASvcA extends ServiceMBeanSupport implements TestHASvcAMBean {
public void startHAService() {
System.out.println("# Starting HA service " + TestHASvcA.class.toString());
}

public void stopHAService() {
System.out.println("# Stopping HA service " + TestHASvcA.class.toString());
}
}

TestHASvcAMBean.java

package test.ha;

import org.jboss.system.ServiceMBean;

public interface TestHASvcAMBean extends ServiceMBean {
}

TestHASvcB.java

package test.ha;
import org.jboss.system.ServiceMBeanSupport;
/**
* Make sure to use the naming standard for MBeans (If the
* source class is named Serious, then the interface must be named
* SeriousMBean).
*
* @author ws
*/
public class TestHASvcB extends ServiceMBeanSupport implements TestHASvcBMBean {
public void startHAService() {
System.out.println("#Starting HA service " + TestHASvcB.class.toString());
}

public void stopHAService() {
System.out.println("#Stopping HA service " + TestHASvcB.class.toString());
}
}

TestHASvcBMBean.java

package test.ha;

import org.jboss.system.ServiceMBean;

public interface TestHASvcBMBean extends ServiceMBean {
}

TestHASvcC.java

package test.ha;
import org.jboss.system.ServiceMBeanSupport;
/**
* Make sure to use the naming standard for MBeans (If the
* source class is named Serious, then the interface must be named
* SeriousMBean).
*
* @author ws
*/
public class TestHASvcC extends ServiceMBeanSupport implements TestHASvcCMBean {
public void startHAService() {
System.out.println("#Starting HA service " + TestHASvcC.class.toString());
}

public void stopHAService() {
System.out.println("#Stopping HA service " + TestHASvcC.class.toString());
}
}

TestHASvcCMBean.java

package test.ha;

import org.jboss.system.ServiceMBean;

public interface TestHASvcCMBean extends ServiceMBean {
}

Customize JBoss HA Singleton Service's Selection Policy
Key of dynamic HA service distrbution and migration is customized activation of HA singleton services on selected node at runtime. JBoss needs sort of "HASingletonElectionPolicy" to decide on which node to activate a certain HA singleton service. The default selection policy is "HASingletonElectionPolicySimple", which always activate HA Service on the "0th" node.
To change the default selection behavoir, we need to provide JBoss with a customized HA selection policy, we call it "TestP" selection policy hereafter:

write a "TestPMBean" interface for the "TestP" MBean.
write a "TestP" MBean, which, implements "HASingletonElectionPolicy" and TestPBean.
configure JBoss to apply this selection policy to the singleton service MBeans in "jboss-service.xml"

TestPMBean.java

package test.ha;

import org.jboss.system.ServiceMBean;

public interface TestPMBean extends ServiceMBean {
void setId(String singletonId);
String getId();
}

TestP.java

package test.ha;

import java.util.List;

import org.jboss.ha.framework.interfaces.ClusterNode;
import org.jboss.ha.framework.interfaces.HASingletonElectionPolicy;
import org.jboss.system.ServiceMBeanSupport;

public class TestP extends ServiceMBeanSupport implements
HASingletonElectionPolicy, TestPMBean {

private String id;

@Override
public void setId(String singletonId) {
this.id = singletonId;
}

@Override
public String getId() {
return this.id;
}

// @Override
public ClusterNode elect(List<ClusterNode> arg0) {
System.out.println(" ### list all nodes before doing selection ploicy name: ");

for (ClusterNode tmpNode : arg0) {
System.out.println(" ## node name: " + tmpNode.getName());
}

// int nodeSeg = getNodeSeq(arg0);
ClusterNode selectedNode = getNodeSeq(arg0);
System.out.println(" ## selected node, policy name: "+ selectedNode.getName());

return selectedNode;
}

/*
* This method changes the selection behavior from default to our customized selection
logic.
*
*/
private ClusterNode getNodeSeq(List<ClusterNode> activeNodes) {

System.out.println(" ## getNodeSeq , policy name " + id);
// A, B, C 3 logical nodes
// 3 physical nodes: A->0, B->1, C->2
int nodeCount = activeNodes.size();
if (nodeCount == 3) {
if (this.id.equals("SS_A")) {
return activeNodes.get(0);
}

if (this.id.equals("SS_B")) {
return activeNodes.get(1);
}

if (this.id.equals("SS_C")) {
return activeNodes.get(2);
}
}

if (nodeCount == 2) {
if (this.id.equals("SS_A")) {
return activeNodes.get(0);
}

if (this.id.equals("SS_B")) {
return activeNodes.get(0);
}

if (this.id.equals("SS_C")) {
return activeNodes.get(1);
}
}

if (nodeCount == 1) {
if (this.id.equals("SS_A")) {
return activeNodes.get(0);
}

if (this.id.equals("SS_B")) {
return activeNodes.get(0);
}

if (this.id.equals("SS_C")) {
return activeNodes.get(0);
}
}

// Default active service on the 0th node
return activeNodes.get(0);
}
}

jboss-service.xml

Create a new folder "META-INF" directly in base folder of "HAServiceDynamicLoadMigration" project. Create a new "jboss-service.xml" in "META-INF" with following content:

<?xml version="1.0" encoding="UTF-8"?>
<server>
<mbean code="test.ha.TestHASvcA" name="myexample:service=TestHASvcA"/>
<mbean code="test.ha.TestP"
name="myexample:service=SingletonServiceControllerA,type=ElectionPolicy">
<attribute name="Id">SS_A</attribute>
</mbean>
<mbean code="org.jboss.ha.singleton.HASingletonController" name="myexample:service=SingletonServiceControllerA">
<attribute name="HAPartition">
<inject bean="HAPartition" />
</attribute>
<attribute name="ElectionPolicy">
<inject bean="myexample:service=SingletonServiceControllerA,type=ElectionPolicy" />
</attribute>
<attribute name="Target">
<inject bean="myexample:service=TestHASvcA" />
</attribute>
<attribute name="TargetStartMethod">startHAService</attribute>
<attribute name="TargetStopMethod">stopHAService</attribute>
</mbean>

<mbean code="test.ha.TestHASvcB" name="myexample:service=TestHASvcB"/>
<mbean code="test.ha.TestP" name="myexample:service=SingletonServiceControllerB,type=ElectionPolicy">
<attribute name="Id">SS_B</attribute>
</mbean>
<mbean code="org.jboss.ha.singleton.HASingletonController" name="myexample:service=SingletonServiceControllerB">
<attribute name="HAPartition">
<inject bean="HAPartition" />
</attribute>
<attribute name="ElectionPolicy">
<inject bean="myexample:service=SingletonServiceControllerB,type=ElectionPolicy" />
</attribute>
<attribute name="Target">
<inject bean="myexample:service=TestHASvcB" />
</attribute>
<attribute name="TargetStartMethod">startHAService</attribute>
<attribute name="TargetStopMethod">stopHAService</attribute>
</mbean>

<mbean code="test.ha.TestHASvcC" name="myexample:service=TestHASvcC"/>
<mbean code="test.ha.TestP" name="myexample:service=SingletonServiceControllerC,type=ElectionPolicy">
<attribute name="Id">SS_C</attribute>
</mbean>
<mbean code="org.jboss.ha.singleton.HASingletonController" name="myexample:service=SingletonServiceControllerC">
<attribute name="HAPartition">
<inject bean="HAPartition" />
</attribute>
<attribute name="ElectionPolicy">
<inject bean="myexample:service=SingletonServiceControllerC,type=ElectionPolicy" />
</attribute>
<attribute name="Target">
<inject bean="myexample:service=TestHASvcC" />
</attribute>
<attribute name="TargetStartMethod">startHAService</attribute>
<attribute name="TargetStopMethod">stopHAService</attribute>
</mbean>
</server>

Now the necessay coding is all done, we are ready to deploy and test HA singleton services.

Deployment

Use you IDE to export "HAServiceDynamicLoadMigration" project as "HAServiceDynamicLoadMigration.sar" archive to "farm" folder of node1.

Step by step we will now simulate the process of starting all the nodes in cluster, shutting down nodes one by one till only one node is left alive, and then recover the custer by restarting failed server one after another. In the mean time we will keep an eye on the console output of each node to check if the HA singleton services migrations among nodes are done as expected, and as programmed in the customized HA selection policy.

Step1. start node1

Now startup node1 and check the console output of node1, you will see output like following:
----

12:33:39,351 INFO [GroupMember] I am (127.0.0.1:32902)
12:33:39,352 INFO [GroupMember] New Members : 1 ([127.0.0.1:32902])
12:33:39,353 INFO [GroupMember] All Members : 1 ([127.0.0.1:32902])
12:33:39,382 INFO [STDOUT]

....

12:33:56,251 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcA

12:33:56,309 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcB

12:33:56,337 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcC

----

The above info shows:

JBoss is started in clustered mode, but currently with only one node
As expected, all 3 HA services are started on node1

Step2. start node2

Now start node2 and check the console output of node1 and node2.

node1 console:
---
12:43:12,454 INFO [GroupMember] org.jboss.messaging.core.impl.postoffice.GroupMember$ControlMembershipListener@1a701a8 got new view [127.0.0.1:50418|1] [127.0.0.1:50418, 127.0.0.1:53538], old view is [127.0.0.1:50418|0] [127.0.0.1:50418]
12:43:12,470 INFO [GroupMember] I am (127.0.0.1:50418)
12:43:12,471 INFO [GroupMember] New Members : 1 ([127.0.0.1:53538])
12:43:12,471 INFO [GroupMember] All Members : 2 ([127.0.0.1:50418, 127.0.0.1:53538])
...
12:43:32,899 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
12:43:32,977 INFO [STDOUT] ## node name: 127.0.0.1:1199
12:43:32,977 INFO [STDOUT] ## node name: 127.0.0.1:1299
12:43:32,977 INFO [STDOUT] ## getNodeSeq , policy name SS_A
12:43:32,977 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199
12:43:32,978 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
12:43:32,978 INFO [STDOUT] ## node name: 127.0.0.1:1199
12:43:32,978 INFO [STDOUT] ## node name: 127.0.0.1:1299
12:43:32,978 INFO [STDOUT] ## getNodeSeq , policy name SS_B
12:43:32,986 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199
12:43:33,020 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
12:43:33,021 INFO [STDOUT] ## node name: 127.0.0.1:1199
12:43:33,021 INFO [STDOUT] ## node name: 127.0.0.1:1299
12:43:33,021 INFO [STDOUT] ## getNodeSeq , policy name SS_C
12:43:33,021 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299
12:43:33,022 INFO [STDOUT] # Stopping HA service class test.ha.TestHASvcC

---
Node1 now has detected, that there is a new member joined the cluater. It updated the sorted list for current alive nodes:
node1 -->[0]
node2 -->[1]

While the cluster structure has changed, JBoss HA Controller will try to update the HA Singleton status with help of the configured HA selection policy.
"TestHASvcA" is configured to use HA selection policy "SS-A", and will activate the "TestHASvcA" on the 0th node, i.e. the node1. Since "TestHASvcA" is already active on node1, so no further action will be done.

"TestHASvcB" is configured to use HA selection policy "SS-B", and will activate the "TestHASvcB" on the 0th node, i.e. the node1. Since "TestHASvcB" is already active on node1, so no further action will be done.

"TestHASvcC" configured to use HA selection policy "SS-C", and will activate the "TestHASvcC" on the 1st node, i.e. the node2. So here the "TestHASvcC" is stopped.

node2 console:

---

12:43:12,481 INFO [GroupMember] I am (127.0.0.1:53538)

12:43:12,482 INFO [GroupMember] New Members : 2 ([127.0.0.1:50418, 127.0.0.1:53538])

12:43:12,482 INFO [GroupMember] All Members : 2 ([127.0.0.1:50418, 127.0.0.1:53538])

12:43:12,642 INFO [STDOUT]

...

12:43:32,901 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

12:43:32,902 INFO [STDOUT] ## node name: 127.0.0.1:1199

12:43:32,903 INFO [STDOUT] ## node name: 127.0.0.1:1299

12:43:32,904 INFO [STDOUT] ## getNodeSeq , policy name SS_A

12:43:32,905 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199

12:43:32,967 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

12:43:32,968 INFO [STDOUT] ## node name: 127.0.0.1:1199

12:43:32,969 INFO [STDOUT] ## node name: 127.0.0.1:1299

12:43:32,969 INFO [STDOUT] ## getNodeSeq , policy name SS_B

12:43:32,970 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199

12:43:33,023 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

12:43:33,023 INFO [STDOUT] ## node name: 127.0.0.1:1199

12:43:33,023 INFO [STDOUT] ## node name: 127.0.0.1:1299

12:43:33,023 INFO [STDOUT] ## getNodeSeq , policy name SS_C

12:43:33,023 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299

12:43:33,027 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcC

Node2 joins cluster and detected, there is a already a member in the cluster. It updated the sorted list for current alive nodes:
node1 -->[0]
node2 -->[1]
Since cluster structure has been changed, JBoss HA Controller will try to update the HA Singleton status.
According to the configured selection policy, "TestHASvcA" and "TestHASvcB" should stay on 0th node, i.e. node1, so nothing will be done for these services on node2.
"TestHASvcC" is configured to start on 1st node, i.e. node2, so node2 will started "TestHASvcC"

---

Step3. start node3

Now start node3 and check the console output of node1, node2 and node3.

node1 console:
---
14:52:24,349 INFO [GroupMember] I am (127.0.0.1:36472)
14:52:24,349 INFO [GroupMember] New Members : 1 ([127.0.0.1:42881])
14:52:24,350 INFO [GroupMember] All Members : 3 ([127.0.0.1:36472, 127.0.0.1:44741, 127.0.0.1:42881])
...
14:52:43,676 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
14:52:43,705 INFO [STDOUT] ## node name: 127.0.0.1:1199
14:52:43,706 INFO [STDOUT] ## node name: 127.0.0.1:1299
14:52:43,706 INFO [STDOUT] ## node name: 127.0.0.1:1399
14:52:43,706 INFO [STDOUT] ## getNodeSeq , policy name SS_A
14:52:43,706 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199
14:52:43,752 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
14:52:43,752 INFO [STDOUT] ## node name: 127.0.0.1:1199
14:52:43,752 INFO [STDOUT] ## node name: 127.0.0.1:1299
14:52:43,753 INFO [STDOUT] ## node name: 127.0.0.1:1399
14:52:43,753 INFO [STDOUT] ## getNodeSeq , policy name SS_B
14:52:43,753 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299
14:52:43,757 INFO [STDOUT] # Stopping HA service class test.ha.TestHASvcB
14:52:43,823 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
14:52:43,825 INFO [STDOUT] ## node name: 127.0.0.1:1199
14:52:43,825 INFO [STDOUT] ## node name: 127.0.0.1:1299
14:52:43,826 INFO [STDOUT] ## node name: 127.0.0.1:1399
14:52:43,826 INFO [STDOUT] ## getNodeSeq , policy name SS_C
14:52:43,827 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399
---

Node3 joins cluster. Node1 updates the sorted list for current alive nodes:
node1 -->[0]
node2 -->[1]
node3--> [2]
Since cluster structure has changed, JBoss HA Controller will try to update the HA Singleton status.
According to the configured selection policy, "TestHASvcA" should stay on 0th node, i.e. node1. "TestHASvcB" should be migrated to 1st node, i.e. node2, so this service will be stopped on node1.

node2 console:
---
[127.0.0.1:36472|1] [127.0.0.1:36472, 127.0.0.1:44741]
14:52:24,298 INFO [GroupMember] I am (127.0.0.1:44741)
14:52:24,298 INFO [GroupMember] New Members : 1 ([127.0.0.1:42881])
14:52:24,298 INFO [GroupMember] All Members : 3 ([127.0.0.1:36472, 127.0.0.1:44741, 127.0.0.1:42881])
14:52:43,675 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
14:52:43,676 INFO [STDOUT] ## node name: 127.0.0.1:1199
14:52:43,676 INFO [STDOUT] ## node name: 127.0.0.1:1299
14:52:43,676 INFO [STDOUT] ## node name: 127.0.0.1:1399
14:52:43,676 INFO [STDOUT] ## getNodeSeq , policy name SS_A
14:52:43,676 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199
14:52:43,748 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
14:52:43,749 INFO [STDOUT] ## node name: 127.0.0.1:1199
14:52:43,749 INFO [STDOUT] ## node name: 127.0.0.1:1299
14:52:43,749 INFO [STDOUT] ## node name: 127.0.0.1:1399
14:52:43,749 INFO [STDOUT] ## getNodeSeq , policy name SS_B
14:52:43,749 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299
14:52:43,751 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcB
14:52:43,822 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
14:52:43,822 INFO [STDOUT] ## node name: 127.0.0.1:1199
14:52:43,823 INFO [STDOUT] ## node name: 127.0.0.1:1299
14:52:43,823 INFO [STDOUT] ## node name: 127.0.0.1:1399
14:52:43,823 INFO [STDOUT] ## getNodeSeq , policy name SS_C
14:52:43,823 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399
14:52:43,824 INFO [STDOUT] # Stopping HA service class test.ha.TestHASvcC

---

Node2 updates the sorted list for current alive nodes too:
node1 -->[0]
node2 -->[1]
node3--> [2]
According to the configured selection policy, "TestHASvcB" should be migrated to 1st node, i.e. node2, so this service was stopped on node1 as shown in the console of node1, and is started on node2.
"TestHASvcB" does not belong to node2 any more and will be stopped on node2, and will be started on node3.

node3 console:
---
14:52:24,367 INFO [GroupMember] I am (127.0.0.1:42881)
14:52:24,368 INFO [GroupMember] New Members : 3 ([127.0.0.1:36472, 127.0.0.1:44741, 127.0.0.1:42881])
14:52:24,369 INFO [GroupMember] All Members : 3 ([127.0.0.1:36472, 127.0.0.1:44741, 127.0.0.1:42881])
14:52:24,875 INFO [STDOUT]
....
14:52:43,679 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
14:52:43,680 INFO [STDOUT] ## node name: 127.0.0.1:1199
14:52:43,681 INFO [STDOUT] ## node name: 127.0.0.1:1299
14:52:43,681 INFO [STDOUT] ## node name: 127.0.0.1:1399
14:52:43,682 INFO [STDOUT] ## getNodeSeq , policy name SS_A
14:52:43,682 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199
14:52:43,755 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
14:52:43,756 INFO [STDOUT] ## node name: 127.0.0.1:1199
14:52:43,756 INFO [STDOUT] ## node name: 127.0.0.1:1299
14:52:43,756 INFO [STDOUT] ## node name: 127.0.0.1:1399
14:52:43,756 INFO [STDOUT] ## getNodeSeq , policy name SS_B
14:52:43,757 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299
14:52:43,826 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
14:52:43,827 INFO [STDOUT] ## node name: 127.0.0.1:1199
14:52:43,827 INFO [STDOUT] ## node name: 127.0.0.1:1299
14:52:43,827 INFO [STDOUT] ## node name: 127.0.0.1:1399
14:52:43,827 INFO [STDOUT] ## getNodeSeq , policy name SS_C
14:52:43,827 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399
14:52:43,831 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcC
---

Node3 joins cluster and detected, there is a already 2 members in the cluster. It created a sorted list for current alive nodes:

node1 -->[0]

node2 -->[1]

node3 -->[2]

Since cluster structure has changed, JBoss HA Controller will try to update the HA Singleton status.

According to the configured selection policy, "TestHASvcC" will be taken over to node3 from node2.

Step4. node1 fails

Shutdown node1, and check console output of node2 and node3.

console node2:
---

14:55:37,727 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

14:55:37,810 INFO [STDOUT] ## node name: 127.0.0.1:1299

14:55:37,810 INFO [STDOUT] ## node name: 127.0.0.1:1399

14:55:37,810 INFO [STDOUT] ## getNodeSeq , policy name SS_C

14:55:37,810 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399

14:55:38,000 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

14:55:38,000 INFO [STDOUT] ## node name: 127.0.0.1:1299

14:55:38,000 INFO [STDOUT] ## node name: 127.0.0.1:1399

14:55:38,000 INFO [STDOUT] ## getNodeSeq , policy name SS_B

14:55:38,001 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299

14:55:38,031 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

14:55:38,039 INFO [STDOUT] ## node name: 127.0.0.1:1299

14:55:38,039 INFO [STDOUT] ## node name: 127.0.0.1:1399

14:55:38,039 INFO [STDOUT] ## getNodeSeq , policy name SS_A

14:55:38,040 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299

14:55:38,043 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcA

...

14:55:46,615 INFO [DefaultPartition] I am (127.0.0.1:1299) received membershipChanged event:

14:55:46,615 INFO [DefaultPartition] Dead members: 1 ([127.0.0.1:1199])

14:55:46,616 INFO [DefaultPartition] New Members : 0 ([])

14:55:46,616 INFO [DefaultPartition] All Members : 2 ([127.0.0.1:1299, 127.0.0.1:1399])

---

When node1 is shutdown, node2 updates the sorted list for current alive nodes:

node2 -->[0]

node3 -->[1]

Since for 2 alive nodes, "TestHASvcA" should run on the 0th node. And now according to the updated alive node list, the 0th node is node2, so node2 will start "TestHASvcA".

console node3:
---

14:55:37,880 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

14:55:38,016 INFO [STDOUT] ## node name: 127.0.0.1:1299

14:55:38,016 INFO [STDOUT] ## node name: 127.0.0.1:1399

14:55:38,016 INFO [STDOUT] ## getNodeSeq , policy name SS_C

14:55:38,017 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399

14:55:38,017 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

14:55:38,017 INFO [STDOUT] ## node name: 127.0.0.1:1299

14:55:38,017 INFO [STDOUT] ## node name: 127.0.0.1:1399

14:55:38,017 INFO [STDOUT] ## getNodeSeq , policy name SS_B

14:55:38,017 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299

14:55:38,032 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

14:55:38,033 INFO [STDOUT] ## node name: 127.0.0.1:1299

14:55:38,033 INFO [STDOUT] ## node name: 127.0.0.1:1399

14:55:38,033 INFO [STDOUT] ## getNodeSeq , policy name SS_A

14:55:38,033 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299

...

14:55:46,625 INFO [DefaultPartition] I am (127.0.0.1:1399) received membershipChanged event:

14:55:46,625 INFO [DefaultPartition] Dead members: 1 ([127.0.0.1:1199])

14:55:46,625 INFO [DefaultPartition] New Members : 0 ([])

14:55:46,625 INFO [DefaultPartition] All Members : 2 ([127.0.0.1:1299, 127.0.0.1:1399])

---

Node3 just updated the alive node list. "TestHASvcC" stays on node3.

Step5. node2 fails

When node2 fails now, the node3 will be left as the onyl alive node. As shown in node3's console,

The alive node list contains now only one node, i.e. node3. The services running on node2, are now migrated to node3.

console node3:

---
14:58:10,737 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcB
14:58:10,800 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcA
14:58:12,491 INFO [GroupMember] ....
14:58:14,078 INFO [DefaultPartition] I am (127.0.0.1:1399) received membershipChanged event:
14:58:14,115 INFO [DefaultPartition] Dead members: 1 ([127.0.0.1:1299])
14:58:14,115 INFO [DefaultPartition] New Members : 0 ([])
14:58:14,115 INFO [DefaultPartition] All Members : 1 ([127.0.0.1:1399])
---

Step6. node1 revocers

Start node1 again, check the console output of node1 and node3

console node1:

---

15:01:35,189 INFO [GroupMember] I am (127.0.0.1:45635)

15:01:35,190 INFO [GroupMember] New Members : 2 ([127.0.0.1:42881, 127.0.0.1:45635])

15:01:35,190 INFO [GroupMember] All Members : 2 ([127.0.0.1:42881, 127.0.0.1:45635])

15:01:35,354 INFO [STDOUT]

---------------------------------------------------------

GMS: address is 127.0.0.1:7900 (cluster=MessagingPostOffice-DATA)

---------------------------------------------------------

15:01:49,131 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:01:49,132 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:01:49,132 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:01:49,132 INFO [STDOUT] ## getNodeSeq , policy name SS_A

15:01:49,133 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399

15:01:49,207 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:01:49,207 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:01:49,207 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:01:49,208 INFO [STDOUT] ## getNodeSeq , policy name SS_B

15:01:49,208 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399

15:01:49,265 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:01:49,266 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:01:49,266 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:01:49,266 INFO [STDOUT] ## getNodeSeq , policy name SS_C

15:01:49,266 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199

15:01:49,272 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcC

15:01:49,366 INFO [Http11Protocol] Starting Coyote HTTP/1.1 on http-127.0.0.1-8180

15:01:49,526 INFO [AjpProtocol] Starting Coyote AJP/1.3 on ajp-127.0.0.1-8109

15:01:49,576 INFO [ServerImpl] JBoss (Microcontainer) [5.1.0.GA (build: SVNTag=JBoss_5_1_0_GA date=200905221634)] Started in 1m:51s:721ms

---

When node1 recovers, the alive node list becomes:

node3 -->[0]

node1 -->[1]
With alive node count equals 2, "TestHASvcA" and "TestHASvcB" should run on the 0th node, so they stay on node3. "TestHASvcC" will be migrated to 1st node, i.e. node1. So we see on the console of nod1, "TestHASvcC" is started.

console node3:
---

15:01:18,345 INFO [DefaultPartition] I am (127.0.0.1:1399) received membershipChanged event:

15:01:18,345 INFO [DefaultPartition] Dead members: 0 ([])

15:01:18,345 INFO [DefaultPartition] New Members : 1 ([127.0.0.1:1199])

15:01:18,345 INFO [DefaultPartition] All Members : 2 ([127.0.0.1:1399, 127.0.0.1:1199])

...

15:01:49,126 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:01:49,127 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:01:49,127 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:01:49,127 INFO [STDOUT] ## getNodeSeq , policy name SS_A

15:01:49,127 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399

15:01:49,206 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:01:49,206 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:01:49,206 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:01:49,206 INFO [STDOUT] ## getNodeSeq , policy name SS_B

15:01:49,207 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399

15:01:49,263 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:01:49,270 INFO [STDOUT] # Stopping HA service class test.ha.TestHASvcC

15:01:49,280 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:01:49,281 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:01:49,281 INFO [STDOUT] ## getNodeSeq , policy name SS_C

15:01:49,281 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199

---

Now node3 is still the 0th node in the cluster, but "TestHASvcC" is expected to run on the 1st node, so this service does not belong to node3 any more, and will be shutdown on node3.

Step6. node2 revocers

Start node2 again.

console node1:
---

15:04:44,024 INFO [GroupMember] I am (127.0.0.1:45635)

15:04:44,024 INFO [GroupMember] New Members : 1 ([127.0.0.1:39250])

15:04:44,025 INFO [GroupMember] All Members : 3 ([127.0.0.1:42881, 127.0.0.1:45635, 127.0.0.1:39250])

...

15:05:00,621 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:05:00,687 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:05:00,688 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:05:00,688 INFO [STDOUT] ## node name: 127.0.0.1:1299

15:05:00,688 INFO [STDOUT] ## getNodeSeq , policy name SS_A

15:05:00,688 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399

15:05:00,688 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:05:00,688 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:05:00,688 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:05:00,701 INFO [STDOUT] ## node name: 127.0.0.1:1299

15:05:00,701 INFO [STDOUT] ## getNodeSeq , policy name SS_B

15:05:00,701 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199

15:05:00,860 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcB

15:05:00,860 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:05:00,861 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:05:00,861 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:05:00,861 INFO [STDOUT] ## node name: 127.0.0.1:1299

15:05:00,861 INFO [STDOUT] ## getNodeSeq , policy name SS_C

15:05:00,861 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299

15:05:00,864 INFO [STDOUT] # Stopping HA service class test.ha.TestHASvcC

---

When node2 joins cluster , there will be 3 nodes in the cluster again, but the alive node list is different than that before nodes failures. The list is now sorted differently:

node3 -->[0]

node1 -->[1]
node2 -->[2]
Before node2 joins cluster, "TestHASvcC" was running on node1. Now with alive count bekoming 3, the node1 as 1st node, will only have "TestHASvcB" runs on it. Therefore "TestHASvcC" will be stopped, and "TestHASvcB" will be started on node1.

console node2:
---
15:04:44,038 INFO [GroupMember] I am (127.0.0.1:39250)
15:04:44,038 INFO [GroupMember] New Members : 3 ([127.0.0.1:42881, 127.0.0.1:45635, 127.0.0.1:39250])
15:04:44,039 INFO [GroupMember] All Members : 3 ([127.0.0.1:42881, 127.0.0.1:45635,
...
15:05:00,624 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
15:05:00,625 INFO [STDOUT] ## node name: 127.0.0.1:1399
15:05:00,625 INFO [STDOUT] ## node name: 127.0.0.1:1199
15:05:00,625 INFO [STDOUT] ## node name: 127.0.0.1:1299
15:05:00,625 INFO [STDOUT] ## getNodeSeq , policy name SS_A
15:05:00,625 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399
15:05:00,732 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
15:05:00,733 INFO [STDOUT] ## node name: 127.0.0.1:1399
15:05:00,733 INFO [STDOUT] ## node name: 127.0.0.1:1199
15:05:00,733 INFO [STDOUT] ## node name: 127.0.0.1:1299
15:05:00,733 INFO [STDOUT] ## getNodeSeq , policy name SS_B
15:05:00,733 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199
15:05:00,863 INFO [STDOUT] ### list all nodes before doing selection ploicy name:
15:05:00,863 INFO [STDOUT] ## node name: 127.0.0.1:1399
15:05:00,863 INFO [STDOUT] ## node name: 127.0.0.1:1199
15:05:00,863 INFO [STDOUT] ## node name: 127.0.0.1:1299
15:05:00,863 INFO [STDOUT] ## getNodeSeq , policy name SS_C
15:05:00,863 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299
15:05:00,868 INFO [STDOUT] # Starting HA service class test.ha.TestHASvcC

---

Node2 joins the cluster as the 2nd node, element [2] of the alive node list, will only have "TestHASvcC" started on it.

console node3:
---

[127.0.0.1:42881, 127.0.0.1:45635]

15:04:44,019 INFO [GroupMember] I am (127.0.0.1:42881)

15:04:44,019 INFO [GroupMember] New Members : 1 ([127.0.0.1:39250])

15:04:44,019 INFO [GroupMember] All Members : 3 ([127.0.0.1:42881, 127.0.0.1:45635, 127.0.0.1:39250])

...

15:05:00,621 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:05:00,689 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:05:00,689 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:05:00,689 INFO [STDOUT] ## node name: 127.0.0.1:1299

15:05:00,689 INFO [STDOUT] ## getNodeSeq , policy name SS_A

15:05:00,689 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1399

15:05:00,704 INFO [STDOUT] # Stopping HA service class test.ha.TestHASvcB

15:05:00,731 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:05:00,732 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:05:00,788 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:05:00,788 INFO [STDOUT] ## node name: 127.0.0.1:1299

15:05:00,788 INFO [STDOUT] ## getNodeSeq , policy name SS_B

15:05:00,789 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1199

15:05:00,789 INFO [STDOUT] ### list all nodes before doing selection ploicy name:

15:05:00,789 INFO [STDOUT] ## node name: 127.0.0.1:1399

15:05:00,789 INFO [STDOUT] ## node name: 127.0.0.1:1199

15:05:00,789 INFO [STDOUT] ## node name: 127.0.0.1:1299

15:05:00,789 INFO [STDOUT] ## getNodeSeq , policy name SS_C

15:05:00,789 INFO [STDOUT] ## selected node, policy name: 127.0.0.1:1299

---

Before node2 joins cluster, node3 was the 0th node in cluster and had "TestHASvcA" and "TestHASvcB"

running on it.

Now the cluster has 3 nodes, node3 will only have "TestHASvcA", running. "TestHASvcB" will be stopped on it, and taken over by node1 as shown in the console of node1.

CONGRATULATIONS!
Now we are done with our expertment. The HA singleton services are again distrubuted on the 3 nodes.

Dynamic HA singeton services distribution and migration is useful, when you have several services which:

each service should run only once in the cluster
each service should be high available, .i.e. could survive node failures
all services not running on one node , but distributed among all node for, for example performance reason.

You might ask, in case of "client-server" system, what if a client is connected to a node, but the required service just runs on another node? We will come to this topic later...

HA Singleton with JBoss 7 Cluster

2012-10-03T12:13:00.000-07:00

Document Version 1.0

Keywords:
JBoss 7 HA Service, HA Singleton, JBoss 7 Cluster, load migration, load distribution, HA Service deployment

Start 2 standalone JBosss 7 server in clustered mode [Draft]

Download JBoss AS 7.1.1.Final and extract the archive to a tmp location, lets say " JBoss_AS_7_1_1_Final"

Create 2 new folders:

jboss7_node1

jboss7_node2

Copy all folders and files under "JBoss_AS_7_1_1_Final" into jboss7_node1.

Copy all folders and files under "JBoss_AS_7_1_1_Final" into jboss7_node2.

Now we have 2 JBoss nodes ready to be started.

Start "jboss7_node1" in clusted mode using command:

./standalone.sh -Djboss.node.name=node1 --server-config=standalone-ha.xml

The node1 is started correctly in clusted mode when following line is shown in console:

[org.jboss.modcluster.advertise.impl.AdvertiseListenerImpl] (MSC service thread 1-3) Listening to proxy advertisements on 224.0.1.105:23364

....

When you are familiar with older JBoss (4x, 5x version), you may expect to see log info about cluster information or node information. But JBoss 7 will not show such info yet.

Start "jboss7_node2" in clusted mode using command:

./standalone.sh -Djboss.node.name=node2 --server-config=standalone-ha.xml -Djboss.socket.binding.port-offset=100

The node2 is started correctly in clustered mode when following line is shown in console:

[org.jboss.modcluster.advertise.impl.AdvertiseListenerImpl] (MSC service thread 1-3) Listening to proxy advertisements on 224.0.1.105:23364

....

Here we still can not see any log info to clusters and nodes.

You could just write a samll "Hello World" web applicaiton and try to deploy it in cluster.
I will just take the "TestWebSec20" web application, add "<distributable/>" to the web.xml file and try put the "TestWebSec20.war" into the folder "deployments" of jboss7_node1.

The console of node1 shows:

21:19:36,185 INFO [org.jboss.as.clustering.impl.CoreGroupCommunicationService.web] (MSC service thread 1-1) JBAS010206: Number of cluster members: 1
....
21:19:37,378 INFO [org.jboss.as.server] (DeploymentScanner-threads - 2) JBAS018559: Deployed "TestWebSec20.war"

Now cluster info is shown, but where is the node2?

Now try to deploy the "TestWebSec20.wa" in node2, put the war file in the folder "deployments" of jboss7_node2.
The console of node2 shows:

21:26:35,227 INFO [org.jboss.as.clustering.impl.CoreGroupCommunicationService.web] (MSC service thread 1-3) JBAS010206: Number of cluster members: 2
....
21:26:37,207 INFO [org.jboss.as.server] (DeploymentScanner-threads - 1) JBAS018559: Deployed "TestWebSec20.war"

Lets take look again the console of node1:

21:26:31,272 INFO [org.jboss.as.clustering.impl.CoreGroupCommunicationService.lifecycle.web] (Incoming-1,null) JBAS010247: New cluster view for partition web (id: 1, delta: 1, merge: false) :

[node1/web, node2/web]

21:26:31,326 INFO [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (Incoming-1,null) ISPN000094: Received new cluster view: [node1/web|1] [node1/web, node2/web]

Now lets shutdown node1, and console of node2 shows:

21:31:10,903 INFO [org.jboss.as.clustering.impl.CoreGroupCommunicationService.lifecycle.web] (Incoming-6,null) JBAS010247: New cluster view for partition web (id: 2, delta: -1, merge: false) : [node2/web]

21:31:10,961 INFO [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (Incoming-6,null) ISPN000094: Received new cluster view: [node2/web|2] [node2/web]

HA Singleton, Cluster Wide Singleton as MBean in JBoss 5, part 3/3

2012-10-02T01:23:00.001-07:00

Document Version 1.0

Keywords:
JBoss HA Service, HA Singleton, JBoss Cluster, load migration, load distribution, HA Service deployment

In "HA Singleton, Cluster Wide Singleton as MBean in JBoss 5", part1-2, we have successfully configured a 2-nodes JBoss cluster, and have reated a "TestHASingleton" MBean. This bean is ready to be deployed as a JBoss HA singleton. If you have followed the part2, you should now have a "TestHASingleton.sar" file. In this part 3 of the serial, we will deploy the "TestHASingleton" Mbean in JBoss cluster. We will also do some experiments to make sure it is really a HA singleton, i.e. it runs ONLY once in the cluster.

Deployment of "TestHASingleton" MBean

Put the "TestHASingleton.sar in "JBOSS_HOME/server/node1/farm/".

You will see following information in the console of node1 :

10:28:09,752 INFO [STDOUT] ### Starting JbHaSingletonSvcSample Singleton Service..

From the above text we can see that the "startSingletonService" method of "JbHaSingletonSvcSample" is called, i.e. the singleton is started on node1.

We check the console output of node2, nothing about the singleton is shown. This is correct.

Shutdown node1

Now, we will shut down node1. Now we will see following info in the console of node2:

10:40:05,383 INFO [STDOUT] ### Starting JbHaSingletonSvcSample Singleton Service..

Find out service on which node will the singleton be active?

step1. shutdown node1 and node2

step2. remove the "TestHASingleton.sar" from farms of both nodes

step3. start node1

step4. start node2

step5. deploy "TestHASingleton.sar" in farm of node2.

The console of node2 shows:

10:48:34,880 INFO [STDOUT] ### Starting JbHaSingletonSvcSample Singleton Service..

10:48:37,657 INFO [STDOUT] ### Stopping JbHaSingletonSvcSample Singleton Service..

The console of node1 shows:

10:48:37,658 INFO [STDOUT] ### Starting JbHaSingletonSvcSample Singleton Service..

This is because, via default, a HA singleton will only active on the "master" node of a cluster (JBoss mainains a sorted list of all active nodes, sorted by the time it joins the cluster. The first node in the list will be choosen as master... ). A HA singleton is active only on the MASTER node.

We could also customize the HA singleton behavior when deciding service on which node should be made active. In such case customized "HA selection policy" is needed...

HASingleton, Cluster Wide Singleton as MBean in JBoss 5, part2/3 part1/3

HA Singleton, Cluster Wide Singleton as MBean in JBoss 5, part 2/3

2012-10-02T01:06:00.005-07:00

Document Version 1.0

Keywords:
JBoss HA Service, HA Singleton, JBoss Cluster, load migration, load distribution, HA Service deployment

Write and deploy an JBoss Mbean

We will write a JBoss MBean called "JbHaSingletonSvcSample"

Create a simple Java project "TestHASingleton" in Eclipse.

An MBean needs an interface, an implementation class, and a "jbosss-service.xml" file for MBean description/deployment.

The interface:

package test.ha;

import org.jboss.system.ServiceMBean;

public interface JbHaSingletonSvcSampleMBean extends ServiceMBean{

}

The implementation class:

package test.ha;
import org.jboss.system.ServiceMBeanSupport;
/**
* The service itself shoul not be written as singleton.
* @author ws
*
*/
public class JbHaSingletonSvcSample extends ServiceMBeanSupport implements
JbHaSingletonSvcSampleMBean {
// The lifecycle
public void startSingletonService() throws Exception {
System.out.println("### Starting JbHaSingletonSvcSample Singleton Service..");
}
public void stopSingletonService() throws Exception {
System.out.println("### Stopping JbHaSingletonSvcSample Singleton Service..");
}
}

Create a "META-INF" folder directly under project root:

TestHASingleton/

src/

META-INF/

Create a "jboss-service.xml" file in "META-INF" folder with following content:

<?xml version="1.0" encoding="UTF-8"?>

<attribute name="TargetStartMethod">startSingletonService</attribute>

<attribute name="TargetStopMethod">stopSingletonService</attribute>

</mbean>

</server>

MBean will be deployed as ".sar" archive. An ".sar" is nothing else that a ".jar" file. To make an ".sar" file, we just need to export the project binaries as "jar" file with Eclipse, i.e. "TestHASingleton.jar",

and then rename it to "TestHASingleton.sar".

HASingleton, Cluster Wide Singleton as MBean in JBoss 5, part3/3 part1/3

HA Singleton, Cluster Wide Singleton as MBean in JBoss 5, part 1/3

2012-10-01T23:52:00.001-07:00

Document Version 1.0

Keywords:
JBoss HA Service, HA Singleton, JBoss Cluster, load migration, load distribution, HA Service deployment

What is HA Singleton?

We know that singletons are the kind of instances or services that exists only once in an application context. When you write a singleton, deploy it in your server, you will get only one instance of this singleton on the server.

The singleton mentioned above is actually class-loader wide singleton, i.e. one instance per class-loader. In case of high availabile cluster, when the above singleton is deployed in cluster, there will be one singleton instance in each cluster-node.

But there are cases where we need cluster-wide singleton. For example, in a clustered auction system,
bid orders can come from different node, but the process which deals with the final trading settelment should run only once in the whole cluster. When the node on which the singleton service fails, another node will start an singleton service automatically. Such singleton is so called cluster-wide singleton, i.e. HA singleton.

We will now take JBoss as example to show how to implement a HA singleton.

JBoss supports deployment of singleton as HA singleton. There are generally 2 ways to deplyoment HA singleton on JBoss

option1: just put deployment archive under "../deploy-singleton/ ", and the deployment service bekomes a HA singleton. Disadvantages of this way are, no hot-deployment support, in case of node failure, service startup time takes longer...
option 2: deploy service as MBean. MBean support hot deployment. Singleton MBean will be deployment on all nodes, but provides service only on one node.

We take the second option, and show how to deploy a MBean as HA singleton:
We will configure and run a 2-nodes JBoss cluster
We will write a simple MBean.
We will deploy the MBean as HA singlleton.

Configure and run a 2-nodes JBoss cluster

Download JBoss 5.1.0_GA from www.boss.org (http://sourceforge.net/projects/jboss/files/JBoss/JBoss-5.1.0.GA). Extract the file to some location, we will name it JBOSS_HOME hereafter.

In the extracted JBoss directory goto JBOSS_HOME/server/, create 2 new folders directly here:

node1
node2

Copy all files in JBOSS_HOME/server/all into node1.
Copy all files in JBOSS_HOME/server/all into node2.

Now we have a JBoss cluster with two nodes ready to run.

Start node1:

./run.sh -c node1 -Djboss.service.binding.set=ports-01 -Djboss.messaging.ServerPeerID=1

When you have followed the above steps, you will see in the console like these:

09:57:56,551 INFO [GroupMember] I am (127.0.0.1:56944)
09:57:56,551 INFO [GroupMember] New Members : 1 ([127.0.0.1:56944])
09:57:56,551 INFO [GroupMember] All Members : 1 ([127.0.0.1:56944])
09:57:56,556 INFO [STDOUT]
.......

09:58:03,364 INFO [Http11Protocol] Starting Coyote HTTP/1.1 on http-127.0.0.1-8180

09:58:03,377 INFO [AjpProtocol] Starting Coyote AJP/1.3 on ajp-127.0.0.1-8109

09:58:03,382 INFO [ServerImpl] JBoss (Microcontainer) [5.1.0.GA (build: SVNTag=JBoss_5_1_0_GA date=200905221634)] Started in 25s:474ms

The node1 is now started.

Start node2:

./run.sh -c node2 -Djboss.service.binding.set=ports-02 -Djboss.messaging.ServerPeerID=2

You will see in the console lines like these:

|1] [127.0.0.1:56944, 127.0.0.1:55687], old view is null
10:01:57,587 INFO [GroupMember] I am (127.0.0.1:55687)
10:01:57,587 INFO [GroupMember] New Members : 2 ([127.0.0.1:56944, 127.0.0.1:55687])
10:01:57,587 INFO [GroupMember] All Members : 2 ([127.0.0.1:56944, 127.0.0.1:55687])
10:01:57,629 INFO [STDOUT]
.....

10:01:59,745 INFO [Http11Protocol] Starting Coyote HTTP/1.1 on http-127.0.0.1-8280
10:01:59,756 INFO [AjpProtocol] Starting Coyote AJP/1.3 on ajp-127.0.0.1-8209
10:01:59,762 INFO [ServerImpl] JBoss (Microcontainer) [5.1.0.GA (build: SVNTag=JBoss_5_1_0_GA date=200905221634)] Started in 17s:400ms

The second node is started, and it joined the cluster.

HA Singleton, Cluster Wide Singleton as MBean in JBoss 5, part2/3 part3/3

SSL Offloading with mod_jk part 6

2012-09-09T12:48:00.001-07:00

Document Version 1.0

Configure Tomcat to accept SSL handling of mod_jk

We will now configure "SSL Termination" for "TestWebSec20" application.
We need to:

1. Generate a self-signed SSL certificate.
This includes creating a ".key" file, a ".csr" file and ".crt file".
2. Configure Apache virtual host i.e. host "ahaha.com" to use SSL
3. Configure Tomcat to accept SSL handling of Apache and mod_jk.

6.1 Generate a self-signed SSL certificate

The following link is the best Tutorial I have ever found for creating self-signed SSL certificate:
http://www.akadia.com/services/ssh_test_certificate.html

Please follow the link to create 3 files in following folders:

"/etc/apache2/ssl.key/ahaha.com.key", here "ahaha.com.key" is the key file name.
"/etc/apache2/ssl.csr/ahaha.com.csr", here "ahaha.com.csr" is the csr file name.

"/etc/apache2/ssl.crt/ahaha.com.crt", here "ahaha.com.crt" is the crt file name.

6.2 Configure Apache virtual host for SSL

Configure Apache virtual host for "ahaha.com" to use SSL.
Add a "vhost-ssl.conf" file to Apache's "vhost.d" folder, and add following content to the file:

####START vhost-ssl.conf
NameVirtualHost *:443
<IfDefine SSL>
<IfDefine !NOSSL>

<VirtualHost *:443>
# General setup for the virtual host
DocumentRoot "/srv/www/vhosts/ahaha.com"
ServerName www.ahaha.com
#ServerAdmin webmaster@example.com
ErrorLog /var/log/apache2/error_log
TransferLog /var/log/apache2/access_log

# SSL Engine Switch:
# Enable/Disable SSL for this virtual host.
SSLEngine on

# SSL protocols
# Supporting TLS only is adequate nowadays
SSLProtocol all -SSLv2 -SSLv3

# SSL Cipher Suite:
SSLCipherSuite ALL:!aNULL:!eNULL:!SSLv2:!LOW:!EXP:!MD5:@STRENGTH

# Server Certificate:
SSLCertificateFile /etc/apache2/ssl.crt/ahaha.com.crt

# Server Private Key:
SSLCertificateKeyFile /etc/apache2/ssl.key/ahaha.com.key

CustomLog /var/log/apache2/ssl_request_log ssl_combined
JkMountCopy On
JkMount / worker1
JkMount /* worker1

</VirtualHost>

</IfDefine>
</IfDefine>
####END vhost-ssl.conf

With this vhost configuration for SSL, a SSL request to "ahahacom" will be first processed by mod_jk. mod_jk will take care of the SSL communication:

providing client with the corresponding SSL certifiacte
do the SSL hand shake
do encryption and decryption
and forward it to Tomcat

The communication between mod_jk and Tomcat are in "plain" text, NO SSL.

6.3 Configure Tomcat to accept SSL handling of mod_jk.

With the SSL configuration in "vhost-ssl.conf", mod_jk will take care of the SSL conmunication to browser, and forward client request to Tomcat. Tomcat need to be configured to believe the request from mod_jk is "secured".

Remembe in part5, we configured tomcat for a "ajp connector".

...
<Connector port="8009" protocol="AJP/1.3" redirectPort="8443"/>
...

To make Tomcat accept mod_jk SSL handling is easy, we just need to change the redirectPort of the above ajp connector to 433:

...
<Connector port="8009" protocol="AJP/1.3" redirectPort="443"/>
...

No restart Apache and Tomcat. Try accessing "secure/HalloSec" again:
http://ahaha.com/TestWebSec20/secure/HalloSec

Your browser will be redirected to HTPPS, and now the server certificate is shown corectly, and SSL port is not shown any more:

part1 part2 part3 part4 part5

SSL Offloading with mod_jk part 5

2012-09-09T12:22:00.001-07:00

Document Version 1.0

Copyright © 2012-2013 beijing.beijing.012@gmail.com

Keywords:
SSLOffloading SSL-Offloading, SSL Termination, Apache, Tomcat, mod_jk configuration, multiple vhosts, multiple SSL certificates one ip

SSL Port Problem

In part4 of this serial, we have successfully configured, a virtual host "ahaha.com", we have configured Apache and mod_jk, so that when a request comes form browser(clien), mod_jk will try to communicate this request to Tomcat. mod_jk talks "ajp language" to Tomcat. Now we need to configure Tomcat so that, he can also understand "ajp".
This could be accmplished by uncommenting following lines of "server.xml" of tomcat:

...
<Connector port="8009" protocol="AJP/1.3" redirectPort="8443"/>
...

Now restart Tomcat and try accessing the "TestWebSec20"applicaiton again:

"http://www.ahaha.com/TestWebSec20/HalloNormal

And now you will see:

Now you could access the "TestWebSec20" web applicaiton through a "real" domain name.
And you do NOT see the port 8080 in browser input any more!

Let's try accessing the secured resource, namly:

http://www.ahaha.com/TestWebSec20/secure/HalloSec

And you will see in you browser:

Cool! Isn't it?

The brower was redirected to HTTPS, and after confirmation of th SSL certificate, we see the protected resource!

But, wait a minute! There is still something wrong!

The browser was redirected to HTTPS, but we see now another port numer 8443 (which is configured for Tomcat https), but we actually don't want to show users the port number in browser. We are expecting something like this

https:///www.ahaha.com/TestWebSec20/secure/HalloSec

When we take a closer look at the SSL certificate details, we will see that the certificate here is the certificate of Tomcat, not the certificate of "www.ahaha.com".

To sum it up, we do NOT want to show the SSL port to brower; we want to show the user our "real" certificate that we created for domain "ahaha.com".

To solve these two problems, we need to introduce a new "term", i.e. so called "SSL offload" or "SSL Termination". This is actually the main purpose of the "SSL Offloading with mod_jk" serial I am writing!

So what is SSLTermination / Offloading?

SSL Termination is used take care of everything about SSL in a centralized position in a system, at the border between secured and unsecured areas. "SSL Terminator" takes care of the SSL things between client and Servers, and all the traffic behinde "SSL Terminator" is not encrypted but thought to be secure.

To make it simple, in our case we want apache / mod_jk to take care of the SSL handling, and by NOT Tomcat.

SSL Offloading with mod_jk part 6
part1 part2 part3 part4

SSL Offloading with mod_jk part 4

2012-09-09T10:33:00.001-07:00

Document Version 1.0

Fronting Tomcat with Apache and mod_jk

Now, we will try to front Tomcat with Apache and mod_jk. The target OS is Linux -OpenSuse12.1. The procedure might be slightly different between different Linux systems.

4.1 Install Apache2

Normmally Apache2 is included in OpenSuse12.1 package. If Apache is installed , there will be an apache2 folder under "etc":
/etc/apache2/

When you can't find this folder, just try to install Apache2 and mod_jk with "Yast".
After installation, start apache2:

# cd /etc/init.d
# ./apache2 start

Now when you type following URL in brower input "http://localhost", you will land on the apache2 default page.

Just now we have also installed mod_jk ( Yast, search mod_jk, and install). But mod_jk is not loaded by apache via default.

4.2 Configure Apache to load/use mod_jk

Configuration of Apache to use mod_jk, three things need to be done:

1. Add a file "mod_jk. conf" to Apache. This will load mod_jk module and specify "worker.properties" file for mod_jk
2. Add a "worker.properties" file to configure mod_jk workers
3. Add vritual host which will use mod_jk

4.2.1 mod_jk.conf

Create a "mod_jk.conf" file in folder "/etc/apache2/conf.d/".
You could also name the file "sample.sss.conf", but important is, the file name must have ".conf" at the end, and the is put in "conf.d" folder. In this way, the file will be found and loaded by apache.

Content of the "mod_jk.conf" file:

#### START mod_jk.conf
# mod_jk configuration for Apache
# Load mod_jk module
LoadModule jk_module /usr/lib/apache2/mod_jk.so

# Tell Apache where to find workers.properties. We assume Tomcat runs on a differerent machine than # Apache. and put workers.properties file near to apache
JkWorkersFile /etc/apache2/conf.d/workers.properties

# mod_jk log configuration
JkLogFile /var/log/apache2/mod_jk.log
JkLogLevel debug
jkLogStampFormat "[%a %b %H:%M:%S %Y]"

# JkOptions indicate to send SSL KEY SIZE,
JkOptions +ForwardKeySize +ForwardURICompat -ForwardDirectories

# JkRequestLogFormat set the request format
JkRequestLogFormat "%W %V %T"

# Send everything for context /TestWebSec20 to worker ajp13
JkMount /TestWebSec20 worker1
JkMount /TestWebSec20/* worker1

# Send everything for context /sampple to worker ajp13
#JkMount /sample worker1
#JkMound /sample/* worker1

##### END mod_jk.conf

Explanation to the "mod_jk.conf" file

1. LoadModule tells Apache to load mod_jk module.
2. JkWorkersFile specifies the worker file location. Workers file tells mod_jk, where to find the real
application (i.e. ip and port of the Server / application).
3. JkMount, have 2 entries, one with another without "*"

4.2.2 workers.properties file

Create a "workers.properties" file in conf.d folder with following content:

##### SATART works.properties
# Define a worker named "worker1"
# Several worker names are separated by ","
worker.list=worker1

# Set properties for worker1 to use ajp13 protocol and run on port 8009
worker.worker1.type=ajp13
worker.worker1.host=localhost
worker.worker1.port=8009
worker.worker1.lbfactor=50
worker.worker1.cachesize=10
worker.worker1.cache_timeout=600
worker.worker1.socket_keepalive=1
worker.worker1.socket_timeout=300

##### END works.properties

Explanation to the workers.properties file:

1. We just configured one worker. In case of more workers, worker names are separated by ",".
For example:
worker.list=worker1, worker2
2. Workers are configured to use / communicate to certain host and port using "ajp" protocol.
When a request comes to apache /mod_jk via http or https, mod_jk will redirect the request NOT to
HTTP or HTTPS ports, but to AJP ports.(We will configure Tomcat to use AJP connector).

4.2.3 Create virtual host and configure the virtual host to use mod_jk

Assume we will configure a new virtual host "ahahacom", two steps are needed:

   Step 1.
Add virtual host to "hosts" file. Edit "hosts" file under "/etc/", add following lines:
127.0.0.1  ahaha.com www.ahaha.com

Step 2.
Create a vhost.conf file under "/etc/apache2/vhost.d/" with following content

#### STAART vhost.conf
<VirtualHost *:80>
ServerAdmin info@ahaha.com
ServerName ahaha.com

# DocumentRoot: The directory out of which you will serve your
# documents. By default, all requests are taken from this directory, but
# symbolic links and aliases may be used to point to other locations.
DocumentRoot /srv/www/vhosts/ahaha.com

# if not specified, the global error log is used
ErrorLog /var/log/apache2/ahaha.com_error.log
CustomLog /var/log/apache2/ahaha.com_access.log combined

JkMount / worker1
JkMount /* worker1
</VirtualHost>
#### END vhost.conf

With vhost configured, the "TestWebSec20" web application could be accessed "later" using following URL:

http://www.ahaha.com/TestWebSec20

Now restart apache:

./apache2 restart

Try accessing the "http://www.ahaha.com/TestWebSec20/HalloNormal" with your browser.

The browser will show an error "Service Temporarily Unavailable"

Check the mod_jk error log we just configured "
/var/log/apache2/ahaha.com_error.log"

You will see a new error log entry like this:

[Mon Sep 03 20:36:32 2012] [error] [client 127.0.0.1] (2)No such file or directory: cannot access type map file: HTTP_SERVICE_UNAVAILABLE.html.var

When you see this error, your virtual host configuration and mod_jk configuration at Apache/mod_jk side are correct!

This above error says that mod_jk can not find the worker, i.e. it can not find the Tomcat server. Remember, mod_jk tries to talk to Tomcat with "ajp" protocol, to certain host name and port (as configured in "workers.properties"), so the question is now, does Tomcat know about the "ajp" thing?
No not yet!

Configure Tomcat to communicate with mod_jk, using AJP
SSL Offloading with mod_jk part 5
part1 part2 part3 part6

SSL Offloading with mod_jk part 3

2012-09-09T09:54:00.002-07:00

Document Version 1.0

Till now we have successfuly protected our TestWevSec20 Web application with SSL.

The 2 servlets are accessed via following links:

http://localhost:8080/TestWebSec20/HalloNormal

And

http://localhost:8080/TestWebSec20/secure/HalloSec, which is redirected to
https://localhost:8443/TestWebSec20/secure/HalloSec

However, for a real web page, we expect to visit the page via a "real" domain name like "www.sample.com", and NOT localhost. We also don't want browser show http/https ports to user.

So we need to hide the application behinde some proxy/load_balancer... what ever, make the "TestWevSec20" to be accessed using following URLs:

http://www.ahaha.com/TestWebSec20/HalloNormal

And

http://www.ahaha.com/TestWebSec20/secure/HalloSec

For this purpose, we need to front Tomcat with Apache and mod_jk

SSL Offloading with mod_jk part 4
part1 part2 part5 part6

SSL Offloading with mod_jk part 2

2012-09-09T09:33:00.001-07:00

Document Version 1.0

Keywords:
SSLOffloading SSL-Offloading, SSL Termination, Apache, Tomcat, mod_jk configuration, multiple vhosts, multiple SSL certificates one ip

In part1, we have created a sample web application TestWebSec20, and could successfully access the servlet "HalloNormal".

Now we will try to access the "HalloSec" servlet. Since the "HalloSec" servlet is declared to be accessd via URL pattern "/secure/HalloSec". So we put following URL in brower input:

http://localhost:8080/TestWebSec20/HalloSec

We get error like "..can not connect to localhost...". And when we have a look at Tomcat's log, we will see that Tomcat has logged a fatal error at start (and we ignored it in part1):

java.io.FileNotFoundException: /home/xx/.keystore (No such file or directory)

This is because, Tomcat detected that a resource is protected with HTTPS, so it tries to locad the SSL keystore (keystore is needed by JAAS) in the users home location (Linux). But for the default Tomcat instllation there is no keystore file available yet.

Create and configure keystore for Tomcat

Creating keystore

In standard JDK package there is "keytool" under JAVA_HOME/bin/. This keytool could be used to create a keystore for Tomcat/JAAS. Run following conmand:

keytool -genkey -alias localhost -keyalg RSA -keystore /opt/local_keystore/localhost_keystore

Now a keystore file named "localhost_keystore" is created. (with password "test1234", we will need the password to configure Tomcat SSL)

Here instead of using default ".keystore" location, we used customer keystore location and keystore name. Now we need to configure Tomcat's ssl connector in server.xml:

...
<Connector port="8443" protocol="HTTP/1.1" SSLEnabled="true"
maxThreads="150"
clientAuth="false" sslProtocol="TLS"
keystoreFile="/opt/local_keystore/localhost_keystore" keystorePass="test1234"/>
...

Start Tomcat and try accessing the secured servlet again:

Browser is redirected to HTTPS, and we need to confirm the SSL certificate:

Click the left side area of browser input to show the certificate. It is the certificate we just created! Successful!!

SSL Offloading with mod_jk part 3
part1 part4 part5 part6

SSL Offloading with mod_jk part 1

2012-09-08T08:48:00.004-07:00

Document Version 1.0

Table of Contents

Part 1 : Create a simple servlet based Web Application "TestWebSec20"
1.1 System Architect
1.2 Technical Infrastructure
1.3 "HalloNormal" Servlet
1.4 web.xml

Part 2 : Create and configure keystore for Tomcat

Part 3 : Run "HalloNormal" and "HalloSec" Web Application

Part 4 : Fronting Tomcat with Apache and mod_jk
4.1 Install Apache2
4.2 Configure Apache to load / use mod_jk
4.2.1 mod_jk.conf file
4.2.2 workers.properties file
4.2.3 Create virtual hosts and configure the virtual host to use mod_jk

Part 5 : SSL port problem

Part 6 : Configure Tomcat to accept SSL handling of mod_jk
6.1 Generate a self-signed SSL certificate
6.2 Configure Apache virtual host for SSL
6.3 Configure Tomcat to accept SSL handling of mod_jk

There seems alway to be a gap between the people who write applications, and those to take care of the running applications, i.e. the operation guys.
It was surprising to find, how things could be misunderstand, even done wrong, because lack of the whole understanding of both areas, concering topics like applicaiton load balancing, scaling, failover, security, SSL ...

Recently I had the chance again in a project to review the system architecture and security concept, and this made me to decide to "document" some of my experinence and considerations, practices ...., not only about this project, but also back to the years, in the hope to help those who just looking for a solution to some problmes.

Lets begin with SSL Termination / SSL Offloading, and this is the part 1/7

With a simple servlet based Web Applicaion, we will trying to make it run like a real world web application, which has its own domain name(s), which run behind a locad balancer, which has resources protected by SSL, and SSL handled by SSL offloading ...

1.1 System Architect:

When we finished, we will have:

1. A simple Web application with partially ssl protected resources
2. Web Applicaiton running in Tomcat, and Tomcat fronted with Apache + mod_jk
3. Web Applicaiton being accessed using "virtual host", with its own SSL certificate
4. Multiple "virtual host" addressing the same Web Application, each having it own certificate

1.2 The necessary technical infrastructure:

1. Linux (openSuse 12.1) with Apache2 installed
2. Tomcat (7.x)
3. Eclispe IDE for Java

Part 1 : Create a simple servlet based Web Application "TestWebSec20"

We will create a servlet based Web Application called "TestWebSec20" with Eclipse IDE.
TestWebSec20 Web Application has 2 servlets:
"HalloNormal" servlet
"HalloSec" servlet. "HalloNormal" will be protected with SSL.

1.3 "HalloNormal" servlet:

import java.io.IOException;
import java.io.PrintWriter;

import javax.servlet.ServletException;
import javax.servlet.http.HttpServlet;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;

/**
* Servlet implementation class HalloSec
*/
public class HalloSec extends HttpServlet {
private static final long serialVersionUID = 1L;

/**
* @see HttpServlet#HttpServlet()
*/
public HalloSec() {
super();
// TODO Auto-generated constructor stub
}

/**
* @see HttpServlet#doGet(HttpServletRequest request, HttpServletResponse response)
*/
protected void doGet(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {
doPost(request, response);
}

/**
* @see HttpServlet#doPost(HttpServletRequest request, HttpServletResponse response)
*/
protected void doPost(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {
// TODO Auto-generated method stub
PrintWriter writer =response.getWriter();
writer.write("Hallo Sec");
writer.flush();
writer.close();
}
}

1.4 web.xml:

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE web-app PUBLIC "-//Sun Microsystems, Inc.//DTD Web Application 2.2//EN" "http://java.sun.com/j2ee/dtds/web-app_2_2.dtd">
<web-app id="WebApp_ID">
<display-name>TestWebSec20</display-name>
<servlet>
<servlet-name>HalloNormal</servlet-name>
<display-name>HalloNormal</display-name>
<description>TestWebSec20</description>
<servlet-class>web.sec.HalloNormal</servlet-class>
</servlet>
<servlet>
<servlet-name>HalloSec</servlet-name>
<display-name>HalloSec</display-name>
<description></description>
<servlet-class>web.sec.HalloSec</servlet-class>
</servlet>

<servlet-mapping>
<servlet-name>HalloNormal</servlet-name>
<url-pattern>/HalloNormal</url-pattern>
</servlet-mapping>

<servlet-mapping>
<servlet-name>HalloSec</servlet-name>
<url-pattern>/secure/HalloSec</url-pattern>
</servlet-mapping>

<welcome-file-list>
<welcome-file>index.html</welcome-file>
</welcome-file-list>

<security-constraint>
<display-name>Example Security Constraint</display-name>
<web-resource-collection>
<web-resource-name>Secure Area</web-resource-name>
<url-pattern>/secure/*</url-pattern>
<http-method>GET</http-method>
<http-method>POST</http-method>
</web-resource-collection>



<user-data-constraint>
<transport-guarantee>CONFIDENTIAL</transport-guarantee>
</user-data-constraint>
</security-constraint>
</web-app>

Explanation to web.xml

* "HalloSec" servlet will be accessed using url "/secure/HalloSec".
This RUL is protected by "security-constraint".
* "auth-constraint" is commented out, so that here no "login form" is required.
* transport-guarantee CONFIDENTIAL makes the path protected by SSL.
A HTTP request to this path /secure/* will be redirected to HTTPS

Deploy the TestWebSec20 Web Application(TestWebSec20.war) to Tomcat (v.7.x).
Start Tomcat. Now try accessing the HalloNormal servlet using following URL:
http://localhost:8080/TestWebSec20/HalloNormal, You willl now see text "Hallo Normal" in your browser:

SSL Offloading with mod_jk part 2
part3 part4 part5 part6

The Next Milestone

HBase Spring Data Sample

Keywords:Spring Data for Hadoop, NoSQL, HBase, Java, Maven

Table of Contents

1. Create a simple Maven project "spring-hbase" in Eclipse

2. Prepare project's pom.xml file

3. Prepare Spring configuration file

4. Write code accessing HBase

5. Run the sample

ActiveMQ Configuration for persistent Message Delivery

Document Version 1.0

Copyright © 2012-2013 beijing.beijing.012@gmail.com

1. On the Client Side

2. On the Server Side

2.4 Running in "NON_PERSISTENT" mode

HTML5 WebSocket Sample with Java Servlet on Server Side

Document Version 1.0

Copyright © 2012-2013 beijing.beijing.012@gmail.com

1. Create a simple web application "WebSocketSample"

2. MyWebSocketServlet.java

3. MyInbund.java

4. wssockrt.html

5. Export ".war" and deploy it on a running Tomcat

6. The "WebSocket" Header

HTML5 Server-Sent Events Sample with Java Servlet as Event Server

Document Version 1.0

Copyright © 2012-2013 beijing.beijing.012@gmail.com

1. Create a simple web application:

2. Create a 3.X style Java Servlet "SseServer.java"

3. Create a SSE.html file

4. Deploy the Web Application

5. Try accessing the SSE.html use browser

JMX Sample: monitor standalone Application use own JMX Server

1. The MBean

2. A JMX Agent

3. Configuration JMX access:

4. Run MyJMXAgent as Java application:

5. Connect to JMX-Server with jconsole

Create Document with Digital Signature using GPG on Linux

Step 1. Mr. Bozz prepare the original contract, and save the file as "contract.txt" :

Step 2. Mr. Bozz sign the "contract.txt" file with his digital signature

Step3. Mr. Bozz creates his private and public key:

Step4. Mr. Bozz try sign the contract again:

Step5. Mr. Bozz send the "contract.txt.gpg" file to Tomy.

Step6. Tomy tries to verify file "contract.txt.gpg"

Step7. Mr. Bozz exports his public and send it to Tomy

Step8. Tomy imports Mr. Bozz into gpg as trusted key:

Step9. Tomy tries to verify file "contract.txt.gpg" again:

Step10. Tomy extracts the content from "contract.ext.gpg:

Step11. Mr. Bozz send his public key to Jerry

Step12. Jerry signs Mr. Bozz public key

Jerry imports Bozz's public key to his GPG utility:

Step13. Mr. Bozz signe the contract again.import the signed key "bozz_signed.asc"

Step14. Tomy verify the new contract file

Initializing Log4j

JPQL vs. Criteria

Deploy JAX-WS Web Service on Tomcat and JBoss

Keywords:

Tutorial EJB3 Integration Test with Arquillian part4 - JBoss 5 managed container and TestNG

Keywords:JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container, arquilllian TestNG

Tutorial EJB3 Integration Test with Arquillian part3 - JBoss 5 managed container

Keywords:JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container, arquillian JUnit

Tutorial EJB3 Integration Test with Arquillian part2 - JBoss 7 remote container

Keywords:JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container

Tutorial EJB3 Integration Test with Arquillian part1 - JBoss 7 managed container

Keywords:JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container

Tutorial EJB3 Integration Test with Arquillian part1 - JBoss 7 managed container

Hibernate MultipleBagFetchException simultaneously fetch multiple bags

JBoss Dynamic Migration of Distributed HA Singletons among Nodes

Keywords:JBoss HA Service, HA Singleton, JBoss Cluster, JBoss load balancing, load migration, load distribution

Step1. start node1

Step2. start node2

Step3. start node3

Step4. node1 fails

Step5. node2 fails

Step6. node1 revocers

Step6. node2 revocers

HA Singleton with JBoss 7 Cluster

Copyright © 2012-2013 beijing.beijing.012@gmail.com

Keywords:JBoss 7 HA Service, HA Singleton, JBoss 7 Cluster, load migration, load distribution, HA Service deployment

Keywords:
Spring Data for Hadoop, NoSQL, HBase, Java, Maven

Keywords:
JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container, arquilllian TestNG

Keywords:
JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container, arquillian JUnit

Keywords:
JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container

Keywords:
JEE integration test, JEE unit test, arquillian sample, JBoss 5, JBoss 7, EJB test, arquillian dependencies, arquillian enbedded container, managed container, remote container

Keywords:
JBoss HA Service, HA Singleton, JBoss Cluster, JBoss load balancing, load migration, load distribution

Keywords:
JBoss 7 HA Service, HA Singleton, JBoss 7 Cluster, load migration, load distribution, HA Service deployment

Keywords:
JBoss HA Service, HA Singleton, JBoss Cluster, load migration, load distribution, HA Service deployment

Keywords:
JBoss HA Service, HA Singleton, JBoss Cluster, load migration, load distribution, HA Service deployment

Keywords:
JBoss HA Service, HA Singleton, JBoss Cluster, load migration, load distribution, HA Service deployment