JavaOne 2010 is just around the corner! Well, sort of. While JavaOne 2010 doesn't actually happen until almost five months from now (September 19-23), the call for papers has already closed. That means, for people like Sharat Chander, the JavaOne Content Selection Chairperson, JavaOne 2010 really is just around the corner. Tori Wieldt recently interviewed Sharat, and has published JavaOne 2010: Interview with the Sharat Chander, Content Selection Chairperson: There's lots of buzz about the content for JavaOne 2010. I took a moment to talk with Sharat Chander, Principal Product Director in Oracle's Application Grid Product Marketing Team, about the content selection process for JavaOne. Tori asked Sharat "what's your role in relation to JavaOne?" Sharat: For the past four years, I've served as the one of Track Leads assigned to review and grade abstracts received during the Call for Papers(CFP). This year, I'm serving as the overall Chairperson for content selection for JavaOne 2010. Individuals can learn more about this year's Conference at: http://www.oracle.com/us/javaonedevelop/062264.html. With regard to the number of submissions received, Sharat says "the community has definitely given us a challenging job in determining what submissions to approve for the event (a nice problem to have!)" The paper/session selection process is similar to what has existed in recent years. Starting in mid-May, emails will be sent out to people whose submissions have been accepted. Some time in June, the final sessions list will be posted to the JavaOne site. Sharat cites three ways you can keep abreast of JavaOne 2010 news:

• Visit the JavaOne web page often
• Subscribe to the @JavaOneConf Twitter feed
• Join the JaveOne Group on Facebook

Read the full interview for more details. In other Java Today news, Arun Gupta presents TOTD #130: Invoking a OSGi service from a JAX-WS Endpoint - OSGi and Enterprise Java: Sahoo blogged about JAX-WS Web service in an OSGi bundle. This Tip Of The Day (TOTD) provides complete steps to create such an application from scratch. We will create an OSGi service, a JAX-WS compliant Web service as hybrid application, and a JAX-WS Web service client. Lets create an OSGi service first. 1. Create a simple OSGi service as explained in TOTD #36. Generate the maven project as... Dustin Marx compares and contrasts Turn Signals and Coding Conventions: In How Turn Signals Work, Karim Nice writes "Turn signals may be the most underutilized device on a car." I sometimes feel that when I witness drivers who cannot be bothered with indicating their intentions through use of a turn signal. Besides being simple common sense, the importance of using a turn signal to convey intentions to others is documented by multiple groups and individuals including driver safety informational sites, automobile insurance sites, law firms and legal information sites... Coding conventions are much like turn signals in that they are often a preferred and expected form of communication between developers who might not have an easier way to communicate the same information. Unfortunately... David Geary continues his JSF 2.0 article series with JSF 2 fu: Ajax components: Among JSF 2's many new features, two of the most compelling are arguably composite components and Ajax support. But their strength is most apparent when the two are combined, making it easy to implement Ajax-enabled custom components with a minimum of fuss. In this article, I'll show you how to implement an autocomplete component that uses Ajax to manage its list of completion items. In doing so, you'll see how you can integrate Ajax into your own composite components... In the Weblogs, Bruno Ghisi introduces DIME - Dependency Injection for Java ME: My masters colleague, Guilherme Hobold, and his friend, Elinton Machado, have created a very interesting project for helping Java ME applications with dependency injection. Actually, it is not only for Java ME (CLDC), but because of there is no reflection, they have done some magic to make it work in a nice way. The project is called DIME, which means Dependency Injection for Java ME. Basically it has a XML to configure the beans (it depends on kXML lib for parsing), a container to get the beans and the classes have to implemented a Bean interface that provides access to the hash table that is used to set the correct values... Masoud Kalili presents Learning GlassFish v3 Command Line Administration Interface (CLI): Terminals and consoles was one of the earliest types of communication interfaces between a system administrator and the system administration layer. Due to this long time presence, command line administration consoles become one the most utilized administration channel for configuring different software ranging from database engines to router’s embedded operating systems. GlassFish provides several administration channels; one of them is the command line administration interface or CLI from now on. The CLI has many unique features which make it very convenient for command line advocates and new administrators... In the Forums, in the Java SE forum, ksemeks is getting a MidiUnavailableException: Audio Device Unavailable: Hi, so here's the deal. When I try to make a new Sequencer object, java throws this exception: javax.sound.midi.MidiUnavailableException: Audio Device Unavailable Here's the whole error... In the JXTA forum, ariel_ro asks Can someone explain this WARNING: Queue full, dropped one or more elements?: I run a multicast client that sends messages across LAN using JxtaMulticastSocket and Datagrams. When I let the system run for a longer time I get this: Apr 27, 2010 5:13:32 PM net.jxta.impl.util.UnbiasedQueue push3 WARNING:... plato__ posted Java TV - JavaTVapi - javax.tv.service missing: I have installed NetBeans 6.8 and jdk1.6.0_20 and the Java TV api from javatv-developers.dev.java.net. I followed all the instructions and I successfully intergrated the api and the emulator to NetBeans IDE. Althought, when creating a new project, I am... Our Spotlight this week is the NetBeans IDE 6.9 Beta Release: NetBeans 6.9 Beta introduces the JavaFX Composer, a visual layout tool for visually building JavaFX GUI applications, similar to the Swing GUI builder for Java SE applications. Additional highlights include OSGi interoperability for NetBeans Platform applications; support for JavaFX SDK 1.3, PHP Zend framework, and Ruby on Rails 3.0; as well as improvements to the Java Editor, Java Debugger, and issue tracking, and more. NetBeans 6.9 Beta is available in English, Brazilian Portuguese, Japanese and Simplified Chinese, as well as in several community-translated languages. Learn more: http://netbeans.org This week's java.net Poll asks What is the most important enhancement in JavaFX 1.3? The poll will be open through Friday. Our latest Feature Article is The Match Maker Design Pattern - a New Place for the Actions, by Michael Bar-Sinai, which describes how to add actions to a system without modifying business objects, add objects without changing actions, and still keep things reusable. We also just published HTML5 Server-Push Technologies, Part 2 by Gregor Roth; this two-part series explains the new Server-Sent Events and WebSockets API in HTML5. And we're also still featuring Biswajit Sarkar's Using Styles, Themes, and Painters with LWUIT, in which you discover how to use some of the new and enhanced features in LWUIT version 1.3. Current and upcoming Java Events:

• April 26-30, Pasig City, Philippines: JavaEE Training Philippines
• April 29, Istanbul, Turkey: Istanbul JUG April Event
• April 30 - May 2, Reston, VA, US: 2010 Northern Virginia Software Symposium: Spring Edition
• May 3-8: JAX 2010
• May 10-14, Redwood City, CA, US: Android Training
• May 15-17, Redwood City, CA, US: Android Application Development
• May 17-21, Cambridge, MA, US: Boston "Web Services and SOA Programming" 5-day Instructor-led ("face to face") Codecamp By Sang Shin
• May 21-23, St. Louis, MO, US: 2010 Gateway Software Symposium
• June 4-5, Dallas, TX, US: 2010 Lone Star Software Symposium: Dallas Edition
• June 7-11, Redwood City, CA: Java Training
• June 14-17, Denver, CO, US: UberConf 2010
• June 22-25, Pasig City, Philippines: Spring Training Philippines
• July 7-9, Pasig City, Philippines: Struts Training Philippines

Registered users can submit event listings for the java.net Events Page using our events submission form. All submissions go through an editorial review before being posted to the site. Archives and Subscriptions: This blog is delivered weekdays as the Java Today RSS feed. Also, once this page is no longer featured as the front page of java.net it will be archived along with other past issues in the java.net Archive. -- Kevin Farnham O'Reilly Media Twitter: @kevin_farnham

Do you remember the email you got to tell you that your jre had a vulnerability? No? What about the fact that Java 5 is in an end of life phase. Given the time it has taken for Java 7 to appear its has somewhat slowed the normal EOL pace of the Java platforms. You may not know that Java 1.4.2  for business will be supported commercially until April 2018,  which is kind of mind-boggling. Thats like offering support for Windows 95 even now. At some point you should move and I would recommend most developers to be at least on Java 6 as its going to be around for a while. Normally when your team releases your own applications as long as things are working the only reason to change exist underlying or dependent software is either to take advantage of new features or due to forced obsolesence by the provider of that software. Its difficult for an engineer or project lead to provide evidence as to why you need to update an application or what are the steps required to even upgrade if you wanted to. How quickly does your AMI , VMware or Azure image get out of date, who is even checking it? If I look at the Java platform itself and really many of the main platforms, PHP, Ruby, .NET , perl, python etc, the pace of innovation has slowed. There obviously is still growth but its still chasing smaller footprints with apis in quick fixup mode after the first set of developers have released products. My team has created a free service based on some technology we released over 5 years ago to check your Java platform and application and provide recommendations on what would need fixing and what you need to do next. It basically is a 5 minute scan and the reports are available to print out to show your boss. if you are interested head on over to http://quality.spikesource.com .

FESI Research Program Overview FESI is the Free and open source software Enterprise Solutions Institute. We are a research program designed to study tomorrow's internet technologies as a means to teach folks in the local workforce how to use technologies our customers will likely want to implement.  We also perform this research to prepare local engineers with the knowledge and skills to help introduce newer technologies into existing enterprises.  Finally, the integration code we create will be uploaded to java.net and available to everyone under the GPL. To achieve these goals, we will roll out our initial research in three phases: phase one we will complete setting up the research program and introduce members to some of the core technologies we work with, phase two we will construct a web-application using well established technologies, and phase three we will begin replacing the well-established technologies with cutting edge (but proven) technologies. To get involved please email me or send an e-mail to fossesi.md@gmail.com Phase one Phase one is currently under way.  This blog will be where we tell the world how things are going.  Java.net will be where we host our project and source all of our completed code.  Twitter, LinkedIn and Facebook will be used to further help promote our efforts.   We will also use this time to recruit more researchers to FESI. We will invite new members into our program to use the Camel tutorial introducing them to many of the technologies we will be using.  Once we have established all of the administrivia, phase one will be complete.  Phase two Phase two is a planning and rapid implementation process.  We will create a sample application using a well-proven existing technologies. To do this we will:

• implement a schema, triggers, stored procedures and views on a relational database (either MySQL or Oracle),  
• employ a middleware product to serve numerous java servlets, and old-school struts applications,
• use hibernate to act as a bridge between the database and the middleware tier, and
• use the web as our presentation layer. 

Phase two should not last more then a couple of months, as the folks who are interested in FESI will already have experience using these technologies. Phase three Phase three will be spent planning the conversion of the servlets and struts applications into a series of endpoints. To do this we will:  

• replace the struts router with Camel routing and mediation,
• convert all of the EJB's into endpoint applications,
• replace backend RMI with JMS,
• replace the middleware product with Apache Felix,
• replace the relational database with Cassandra, and
• move all stored procedures, triggers and views into Java Pojos.

Conclusion The hands-on approach of actually performing the work on a sample application will teach the skills necessary to fill positions that are opening up at many of our local customers, and will provide the knowledge necessary to help steer the customers in the direction of technologies that work.  Our integration code will be available to the general community, and hopefully will serve as the basis for similar conversions. This is the initial plan subject to changes based on the desires of the research participants. I expect this program it to take the better part of a year to complete. 

My masters colleague, Guilherme Hobold, and his friend, Elinton Machado, have created a very interesting project for helping Java ME applications with dependency injection. Actually, it is not only for Java ME (CLDC), but because of there is no reflection, they have done some magic to make it work in a nice way. The project is called DIME, which means Dependency Injection for Java ME. Basically it has a XML to configure the beans (it depends on kXML lib for parsing), a container to get the beans and the classes have to implemented a Bean interface that provides access to the hash table that is used to set the correct values. The XML file uses the same notation as Spring. Here is some sample code that I got from the project page. XML file (also supports map and other types).                                                                   element1                         element2                                                                         movies2.txt                               Make a bean: import java.util.Hashtable; import org.dime.Bean; import test.collection.FooList; public class Foo extends FooAbstract implements Bean {       private String filename;       private FooList fooSet;       public String getFilename() {             return filename;       }       public void setProperties(Hashtable properties) {             filename = (String)properties.get("filename");             fooSet = (FooList)properties.get("fooSet");       } } Load the XML file and get the bean: BeanContainer container = new ClassPathXmlBeanContainer("file.xml"); Foo foo = (Foo)container.getBean("Foo"); System.out.println(foo.getFilename()); Very nice job! Take a look in the project page for more information. Take a try! http://dime.dev.java.net Cheers, Bruno Ghisi

On April 28th, 2010, we will be kicking off the first phase of FESI's research.  This where we set up the project, and start to get our hands dirty. Attached is the flyer we will use to promote FESI and what we're trying to accomplish.  Hopefully it will help us get the word out to the local community. Also as the kickoff, we'll be participating with a networking group that meets monthly.  I honestly dont' think a ton of folks will sign up for FESI tomorrow night, and that's really not the point.  The kickoff is the start date of FESI.  We can look back on this date in years to come as the birth, the place where it all began, and a date to celebrate our accomplishments in the future. Good luck, and we're looking forward to seeing you there!

This week is my last at Oracle, next week I'll be starting a new job with Mitre. I started looking around for a "Plan B" prior to the Sun acquisition closing in February, mainly due to uncertainty about whether I'd be offerred a position with Oracle. A friend introduced me to an excellent opportunity at Mitre and, after a couple of rounds of interviews and a lot of thought, Plan B morphed into Plan A. I've enjoyed my short stint with Oracle, it was great to be able to work on Jersey again after 18 months or so doing other things and hopefully the declarative hyperlinking extensions and WADL generation improvements I've been working on will have legs after I leave. I don't know yet whether my new role will afford the opportunity to contribute further to Jersey and/or JAX-RS but I'll be working on the application of open services technologies to national problems so its possible. I plan to continue blogging here when I can, so stay tuned for the next chapter.

Content available at: http://blogs.sun.com/arungupta/entry/totd_130_invoking_a_osgi

We've just published Michael Bar-Sinai's new article, The Match Maker Design Pattern - a New Place for the Actions. In this article, Michael tackles some of the problems associated with object-oriented design within the context of the changing requirements that the real-life business world imposes on software applications over time. To see what I mean, here's Michael's opening paragraph: Software systems often deal with similar concepts, whose behavior differs only slightly. Classic Object-Oriented design deals with such cases using inheritance; overriding the calculateSalary() method in different Employee subclasses allows the rest of the application to remain oblivious to the subtle differences between the salary algorithms of Manager, Engineer, and AnnoyingCeoNephew. In the article, Michael presents a solution that, while not a "silver bullet", provides benefits that work in a majority of real-life cases where you're updating software applications to address changed business requirements. His section "Dealing With New Classes On The Fly" opens with: A MatchMaker instance does hold a Map mapping classes to handlers, but with a twist. When it is given an instance of a class for which a handler was not explicitly specified, it embarks on a breadth-first search (BFS) up the class hierarchy, starting from that object's class, looking for a class for which a handler was explicitly specified. A property of BFS is that it ensures that the class found is the closest one to the original object's class. Note that there might be a few classes with the same distance. Figure 1. Class hierarchy and handlers example. The article contains plenty of code snippets that illustrate the match maker technique. Michael concludes the article with this summary: We have seen the Match Maker pattern, which allows programmers to:

• add actions on objects without modifying the objects those actions operate on;
• add objects and have appropriate actions automatically operate on them; and
• work on a subclass while holding only a superclass reference to it, without downcasting.

We have seen two implementations, one simple yet not statically type safe, and one effectively type safe. Creating a full statically type safe solution is left as an exercise to the reader, providing the reader is looking for a subject for a PhD thesis. I have used this this pattern in production code for analyzing emails, drawing tables, building UI, and handling JMS messages; I hope you'll find it useful too. As you can see, not only is The Match Maker Design Pattern - a New Place for the Actions interesting and informative, it's also fun to read! In Java Today, Frank Kieviet, departing manager of the OpenESB project, presents Working with java.net: tips and tricks for project owners: In the past two years I've managed the OpenESB open source project. This project is hosted on java.net. In this post I outline a number of tips and tricks I've learned while managing this project on java.net. This is useful information for my successor, but it also may be useful to other people who are managing projects on java.net. Web presence: An important factor in the success of an open source project is its web presence: it will be the first point of contact between a potential new user and the project. New visitors will want to see answers to these four items immediately... Pelegri presents Fundamo, OSGi, iPad.. and More GlassFish News - April 24rd, 2010: Financial services on the go - GlassFish for Fundamo and profit: Alexis recently published a new Adoption Story on how Fundamo uses GlassFish v2 and OpenMQ for its Enterprise Platform. Overview at stories entry, details in questionnaire, and an overview in this earlier short video interview. We are always interested in more GlassFish adoption stories, both from (non-paying) users and from (paying) customers.   Stories come from all industries and around the world, the last few entries are PSA Peugeot Citroën (France/Auto), iVox (Belgium/Print), NHIH (US/Gov-Health Care) and Suncorp (Australia/Finantial)... James Sugrue continues his design pattern series with Design Patterns Uncovered: The Mediator Pattern: Today's pattern is the Mediator pattern, used to handle complex communications between related objects, helping with decoupling of those objects. Mediator in the Real World: An airport control tower is an excellent example of the mediator pattern. The tower looks after who can take off and land - all communications are done from the airplane to control tower, rather than having plane-to-plane communication. This idea of a central controller is one of the key aspects to the mediator pattern... Geerjan Wielenga continues his Small MultiView Editor Experiment (Part 2): The sample for developers creating multiview editors on the NetBeans Platform (described yesterday in this blog) is now slightly more extensive, since you can add new nodes in the visual view, which will then automatically be added to the source editor (as well as the visual editor, of course)... In the Weblogs, Kohsuke Kawaguchi is Introducing InfraDNA, the Hudson company: As I wrote in my farewell note, I was working on starting a new company around Hudson. It took longer than I initially anticipated, but it's finally open for business! The company will provide two things; one is support, so that I can answer your questions and problem reports in a timely fashion, and the other is consulting, so that I can help you develop custom plugins, or provide on-site support to work on some tricky problems. The name of the company is InfraDNA because I think of Hudson more as an infrastructure on which all kinds of server-side automation/tools can be built/deployed, and because I think this stuff is built into me (as in DNA)... John Ferguson Smart describes Grouping tests using JUnit categories: In a well-organized build process, you want lightning-fast unit tests to run first, and provide whatever feedback they can very quickly. A nice way to do this is to be able to class your tests into different categories. For example, this can make it easier to distinguish between faster running unit tests, and slower tests such as integration, performance, load or acceptance tests. This feature exists in TestNG, but, until recently, not in JUnit. Indeed, this has been missing from the JUnit world for a long time... In the Forums, kschaefe expresses his view in the SwingLabs forum on JXEditorPane honoring display properties: I believe that we should change the default construction of JXEditorPane to honor display properties. This means that when the font or colors are set the rendered documents will honor them. Core added this client property in 1.5, but left the old style... In the Metro and JAXB forum, John Hite asks Where is databinding done?: I'm trying to figure out where metro converts SOAP messages to java objects. Is this done in a Tube? Can anyone point me in the right direction? Thanks, John In the LWUIT forum, rgucci finds that Calling BlackBerry browser to play video blocks the whole device: Developing on the BlackBerry port, OSv5.0, Bold 9700 simulator. I am calling the native BlackBerry Browser to launch the BlackBerry media player to play an mp4 file stored in the SDCard. Everything works and plays, the video and sounds are... Our Spotlight this week is the NetBeans IDE 6.9 Beta Release: NetBeans 6.9 Beta introduces the JavaFX Composer, a visual layout tool for visually building JavaFX GUI applications, similar to the Swing GUI builder for Java SE applications. Additional highlights include OSGi interoperability for NetBeans Platform applications; support for JavaFX SDK 1.3, PHP Zend framework, and Ruby on Rails 3.0; as well as improvements to the Java Editor, Java Debugger, and issue tracking, and more. NetBeans 6.9 Beta is available in English, Brazilian Portuguese, Japanese and Simplified Chinese, as well as in several community-translated languages. Learn more: http://netbeans.org This week's java.net Poll asks What is the most important enhancement in JavaFX 1.3? The poll will be open through Friday. Our latest Feature Article is The Match Maker Design Pattern - a New Place for the Actions, by Michael Bar-Sinai, which describes how to add actions to a system without modifying business objects, add objects without changing actions, and still keep things reusable. We also just published HTML5 Server-Push Technologies, Part 2 by Gregor Roth; this two-part series explains the new Server-Sent Events and WebSockets API in HTML5. And we're also still featuring Biswajit Sarkar's Using Styles, Themes, and Painters with LWUIT, in which you discover how to use some of the new and enhanced features in LWUIT version 1.3. Current and upcoming Java Events:

• April 26-30, Pasig City, Philippines: JavaEE Training Philippines
• April 29, Istanbul, Turkey: Istanbul JUG April Event
• April 30 - May 2, Reston, VA, US: 2010 Northern Virginia Software Symposium: Spring Edition
• May 3-8: JAX 2010
• May 10-14, Redwood City, CA, US: Android Training
• May 15-17, Redwood City, CA, US: Android Application Development
• May 17-21, Cambridge, MA, US: Boston "Web Services and SOA Programming" 5-day Instructor-led ("face to face") Codecamp By Sang Shin
• May 21-23, St. Louis, MO, US: 2010 Gateway Software Symposium
• June 4-5, Dallas, TX, US: 2010 Lone Star Software Symposium: Dallas Edition
• June 7-11, Redwood City, CA: Java Training
• June 14-17, Denver, CO, US: UberConf 2010
• June 22-25, Pasig City, Philippines: Spring Training Philippines
• July 7-9, Pasig City, Philippines: Struts Training Philippines

Registered users can submit event listings for the java.net Events Page using our events submission form. All submissions go through an editorial review before being posted to the site. Archives and Subscriptions: This blog is delivered weekdays as the Java Today RSS feed. Also, once this page is no longer featured as the front page of java.net it will be archived along with other past issues in the java.net Archive. -- Kevin Farnham O'Reilly Media Twitter: @kevin_farnham

          Learning GlassFish v3 Command Line Administration Interface (CLI) Terminals and consoles was one of the earliest types of communication interfaces between a system administrator and the system administration layer. Due to this long time presence,  command line administration consoles become one the most utilized administration channel for configuring different software ranging from database engines to router’s embedded operating systems. GlassFish provides several administration channels; one of them is the command line administration interface or CLI from now on.  The CLI has many unique features which make it very convenient for command line advocates and new administrators which like to get familiar with CLI and use it in the daily basis.  The CLI allows us to manage, administrate, and monitor any resources which application server exposes to the administrators. So we can perform all of our administration tasks just by firing up a terminal, navigating to glassfish bin directory and executing the asadmin script which is either a shell script in UNIX based operating systems or a batch file in Windows operating system. The CLI has some degree of embedded intelligence which helps us to simply find the correct spelling and similar commands to the command or phrase that we try. Therefore with basic knowledge of GlassFish application server administration we can find out the command that we need and proceed with the task list which we have in front.       1 The CLI environment CLI is an administration interface which let administrators and developers to change configurations like changing a listener port; performing life cycle related tasks like deploying a Web application, or creating a new domain; monitoring different aspects of the application server; and finally performing maintenance duties like creating backup from the domain configuration, preparing maintenance reports and so on. Door to the CLI interface is a script which is place in glassfish_home/bin directory. For Linux and UNIX platforms it is a shell script named asadmin and for the Windows platform it is a batch file named asadmin.bat. So, before we start doing anything with the CLI interface we need to either have glassfish_home/bin in the system path or we should be in the same directory which the script resides.   Add glassfish_home/bin to operating system path We can add glassfish_home/bin to windows operating system path by using the following procedure. You should have the glassfish_home variable already defined, if not you can replace %glassfish_home% with the full path to glassfish installation directory.

• Right click on My Computer icon and select properties item
• Select advanced tab in the from the tab set
• Click on environment variables button, you will see two set of variables one for user and one for the system. I suggest you choose the user space to add your variable or update the currently defined variable.
• If you can not find path in the list of variables, create a new variable named path and sets its value to %glassfish_home%/bin. If the variable is present then add ;glassfish_home/bin to the end of the variable’s value.

§     Close all cmd windows and open a new one to have the new path applied And for UNIX and Linux machines, we can add it to operating system path by using the following procedure. Replace "${glassfish_home}" with the full installation of Glassfish. Open a text editor like ktext, gedit or any editor which you are familiar with Open ~/.bashrc file using the file menu At the end of the file (last line) add export PATH="${glassfish_home}"/bin:"${PATH}" Close all terminals and fire up a new terminal console to have the effects applied Info We can execute the asadmin.bat in windows by calling asadmin.bat or asadmin in a cmd windows. We should be either in the bin directory or we should have the script in path. For Linux and UNIX, if we have the script in path we can call asadmin in a terminal window, if we do not have it in the path then we should be in the bin directory and execute it by typing ./asadmin in the terminal window. Info In this article I will use an operating system natural syntax, for example I will use asadmin version which can be interpreted as asadmin.bat version  in windows or ./asadmin version in UNIX or the asadmin version itself when we have asadmin in our path. 2 Getting started with CLI 2.1 The commands format The asadmin script is the door to GlassFish command line interface and all of its commands follow a de-facto naming convention and a standard format which we need to remember. All command names are self explaining and we can understand the command purpose from the command name. All of the asadmin commands use one or multiple words as the command name to completely explain the command purpose. These words which form the command name are separated using a dash (-). The naming convention also provides another learning point about the asadmin commands. Commands are categorized into lifecycle management commands, configuration commands, listing commands and monitoring commands. Each command starts with a verb which determines the command nature, for example: create, delete, update, list, and configure are the starting words for many of asadmin commands.  The command syntax is shown in the following snippet. command-name options operands The command-name can be start-domain, stop-domain, and so on. The options are parameters which affects the command execution, for example the domain instance port. the JDBC connection pool data source class name and so on.  The operand are usually name of the resources which the command may affect, like name of the connection pool that we want to create, name of the domain which we want to start and so on. 2.2 The asadmin operating modes The asadmin utility which we use to execute our administrative commands operates in two modes. Each mode is suitable for specific set of tasks and has its own characteristics when it comes to performance and productivity. Headless operating mode In the headless mode which is very suitable for writing scripts  we call the asadmin  utility and pass the command name along with its options and operands. The asadmin utility launches a JVM and execute the command. The result of the command will be either 0 as failure or 1 as a success. An example of this mode is similar to the following script: asadmin version This will simply execute the version command which shows the application server version. To get help for a command in this mode we can use asadmin help version or we can use asadmin version --help. Multimodes operating mode This operating mode let use execute multiple commands in one JVM instance, therefore it is faster for executing commands. This execution mode is not suitable for script developers as we should execute the commands in asadmin environment and not in the shell command line. To execute a command in this mode we should enter the asadmin environment and then enter our commands along with options and operands. For example: asadmin  A asadmin> version --help B A: Hit enter B: Command name with parameters The asadmin utility is an operating system native shell script which wrap around the actual Java application responsible for executing the administration commands which we issue. Later on we will see how this script can fall useful for administrating some domains from a system which is behind a proxy. 2.3 Local commands Two sets of commands are present in asadmin which help administrators perform variety of tasks. Local commands are a set of commands which either affects the environment which application server is running or it needs accessing the application server environment locally to execute some scripts or batch files to perform a job. A good example of local commands is domain lifecycle management commands which include creating a new domain, deleting a currently available domain, or starting a domain. All of these commands need to either access the application server installation directory layout or need to run a script on the target machine to perform the task. Another place which we can discuss the local commands is the Java DB copy included with the GlassFish application server installation bundle. The following command start the database on the local machine. The command starts the database by creating another Java process directly by using the operating system shell. asadmin start-database You may wonder why this command has not options or operands. Generally all GlassFish administration commands need the minimum possible options and operands to perform the task when we use default setting and configuration.   Each command has its own set of parameters but most of the asadmin commands, either local or remote, have some shared parameters which are listed in table 1. Command’s parameters can have a default value which will be used if we do not specify it. The default value for each parameter is underlined. Table 1 List of shared parameters between local and remote commands Parameter Acceptable values Description --terse or –t false, true Indicates that any output data  must  be  very  concise --echo, -e true, false If set to true, the command-line statement is echoed on the standard output. --interactive, -I true, false If set to true, only the required password options are prompted. --passwordfile Path to the password file We will discuss it in section 5 of this article --user, -u No default value A user with administrative privilege on target domain. These parameters have no effect on the command execution but they just change the way that the command shows us the information and how we interact with the commands. 2.4 Remote commands In the Opposite side of the local commands we have remote commands which form a set of commands that affects the running application server instance configuration and access the application server environment and file system using an application deployed in the application server itself. Therefore the target instance should be running and there should be a network route between the administration workstation and the GlassFish instance running on the server machine. When we execute a remote command we should let the asadmin know where the target application server instance is running and what the instance administration port is.  The asadmin uses this information along with the authentication information to communicate with the application server over HTTP or HTTPs using a REST like URL pattern for commands and parameters. Figure 1 demonstrates the remote and local commands. Figure 1 The asadmin utility communication with remote and local GlassFish domain and installation In figure 1 you can see that in the running GlassFish instance there is an application deployed under /__asadmin/ context and listens for incoming asadmin commands. This application is listening on admin listener which is by default uses port number 4848. Listing 1 shows a sample command content when it is heading toward the application server. It should have already flashed a light in your head about the RESTful Web services development articles or books that you have already read. Listing 1 content of HTTP request for version command GET /__asadmin/version HTTP/1.1 Content-type: application/octet-stream Connection: Keep-Alive Authorization: Basic YWRtaW46YWRtaW5hZG1pbg== User-Agent: hk2-agent     #A      Cache-Control: no-cache Pragma: no-cache Host: localhost:4848 Accept: text/html, image/gif, image/jpeg, *; q=.2, */*; q=.2 #A The asadmin’s agent name Listing 2 shows the response generated by application server for version command. As you can see, the response is plain and simple. Listing 2 response generated by application server for version command HTTP/1.1 200 OK Content-Type: text/html Transfer-Encoding: chunked Date: Mon, 12 Jan 2009 19:26:01 GMT Signature-Version: 1.0 message: GlassFish v3 Prelude (build b28c) exit-code: SUCCESS                                                                                    #A   If you are thinking that trying URLs in your browser will result in some output, you are completely right. For example if you try browsing http://127.0.0.1:4848/__asadmin/uptime you will get a result page similar to figure 2.   Figure 2 Using browsers to run asadmin commands and view the result in browser Now you may have some questions about how a command can generate an HTML oriented output in the browser while it can generates a plain text output when we use asadmin to execute the same command. The answer can be found in the way that browser identifies itself and asadmin utility identifies. The asadmin utility uses hk2-agent as the agent name and asadmin web application respond to this types of agent with a plain text response while it provides a full readable html for other agent types. The only time that we feel difference between these two sets of commands is when we try to execute a local command on a remote machine.  For example creating a domain on a remote machine is not possible unless we have a node agent installed on that machine and we are using DAS to control that machine. A good example of remote commands is application deployment command. We can deploy an application archive like an EAR or WAR file using asadmin. The deployment command will upload the file to the server and server takes care of the deployment. Following command will deploy the gia.war to the default glassfish instance. asadmin deploy gia.war The deploy commands has several parameters, we will discuss these parameters in details in subsequent sections. You may have already asked yourself how we identify which domain we want to execute a command against. Your answer is behind some shared parameters between all remote commands to distinguish the target domain which the command affects. These parameters are listed in table 2 along with the description and default values. Table 2 List of all shared parameters between remote commands Parameter Default values Description --host, -H localhost when no value is determine for this parameter, the asadmin utility will try to communicate with localhost --port, -p 4848 when no value is determine for this parameter, the asadmin utility will try to communicate with the determined host on port 4848 --secure, -s false If we have our asadmin application on a secure listener then we will use this parameter to prevent any security breach. All of these default parameters value’s assumption allows us to have more compact, easy to write and understand commands. Based on the connate of this table we can say that the sample deploy command will deploy the application into a domain running on localhost with port 4848 assigned to its admin listener. 3 Performing common tasks with CLI Application server administration is composed of some common tasks which deal with common concepts of Java EE application servers like domain administration, application lifecycle administration, and managed resources administration. Along with these common tasks there are many other tasks which might be common between application servers or not. These tasks includes listeners administration, resource adapter administration, application server security administration which includes both application server security and Java EE related security definitions, virtual servers administration, monitoring and so on. In this section we will discuss the common administration tasks. 3.1 Domain lifecycle administration Domains are foundation of GlassFish application server instances; each domain is an independent instance of application server which uses the same application server installation and hardware resources. Domain lifecycle administration covers creating, starting, stopping, configuration backup, deleting the domain, and domain configuration recovery using previously created backups. Creating a new domain Creating a new domain is fairly simple when you know what attributes distinguishes one domain from another. A sample command which creates a domain is as follow: create-domain --user admin --adminport 4747  --domaindir /opt/dev/apps/domains/GiADomain --profile developer --instanceport 8282 --domainproperties jms.port=8687 --savemasterpassword=true --passwordfile=/opt/dev/apps/passfile.txt  --savelogin=true GiADomain This is a sample of create-domain command with most of the applicable parameters. We are using --adminport to determine the admin port, --domaindir let us create the domain in none default path like what we used instead of glassfish_home/domains which is the default directory. The --instanceport let us determine the port that default application listener uses. Some of the domain properties like JMS port, JMX port and so on can be determined using the --domainproperties parameter. All parameters are self describing except for following three parameters which you need to understand them thoroughly.

• The --savemasterpassword parameter: Setting this option to true asks the domain creating process to save the master password into a file and facilitate the unattended (headless) domain startup. We will discuss this password in more details in section 4 of this article. For now just remember that default value for master password is changeit. I am sure that value of the password is talking for the importance of changing it.
• The --passwordfile parameter: The asadmin utility policy encourages the administrators not to enter the password in the console and instead use a file which includes the required password. During the domain Creating two passwords are required which are domain administration password and master password. To avoid typing the passwords we can store them in a plain text file similar to the following snippet and pass it to any command that we execute instead of typing the passwords in an interactive way.

AS_ADMIN_PASSWORD=adminadmin AS_ADMIN_MASTERPASSWORD=changeit The file content is in properties file format and can be created using any text editor. If we do not use the --passwordfile parameter and a password file containing the passwords, the asadmin utility will asks for passwords interactively.

• The --savelogin parameter: We will discuss automatic login with more details in section 4 of this article, but to be brief, this parameter make it possible to execute our commands on different domains without providing username and password on every command execution.

Starting and stopping a domain We can start a domain using the start-domain command. An example of starting a domain is similar to the following snippet. start-domain --debug=true  --verbose=true --user admin --passwordfile=/opt/dev/apps/passfile.txt  --domaindir= /opt/dev/apps/domains/  GiADomain              Although the command can be as simple as start-domain domain1 but that is for occasions when we are using all of the default values for any parameter and no customization has happened during the time which we create the domain. The --debug=true means that domain is started in debug mode and we can attach our debugger to the running instance. The --verbose=true means that we want to see all messages that application server emits during the startup; it is useful to catch errors and troubleshoot. The --passwordfile parameter lift the need to enter the admin and master passwords. The file content is similar to password file we used during the domain creation. And finally we need to determine where the domain resides if we create the domain in a non-standard location. To stop a domain we can use a command similar to the following snippet which is fairly simple. stop-domain  --domaindir= /opt/dev/apps/domains/ GiADomain We just need to provide the path to the directory which contains the domain. The only applicable shared parameters are the items that we used. None of the other shared parameters is applicable in this command. Domain utility commands There are two utility commands for domain administration. The first command lists all of the domains which reside in the given directory along with the domain status indicating whether the domain is running or not. And the second command checks the structure of domain.xml file.       list-domains --domaindir=/opt/dev/apps/domains/     verify-domain-xml  --verbose=true --domaindir= /opt/dev/apps/domains/  GiADomain The verify-domain-xml is very useful when we edit the domain.xml file manually. Domain configuration backup and recovery Yes, we can backup a domain configuration and later on restore that backup if we messed up with the domain configuration. Each backup file is an archive which contains description files and vital domain configuration files including domain.xml. By default all backups are placed inside a directory named backups which resides inside the domain directory. The backup file names follow a pattern similar to sjsas_backup_v00001.zip which means that it is the first backup of this domain. To create a backup a command similar to the following snippet will do the trick. backup-domain --description "backup made for glassfish in action book" --domaindir /opt/dev/apps/domains/ GiADomain After some times we will have several backup archives for each of our domains, we will need a command to get a list of all available backups or backups related to one single domain. To get list of backups for a single domain we can use: list-backups --domaindir /opt/dev/apps/domains/ GiADomain Or to get a list of all backups of all domains in the default domains directory we can use list-backups command alone with no parameters. Now that we can list the available backups for each domain we need command which allows us to restore a domain. To restore a domain to one of its previous backups we should stop the domain and then use the restore command which is similar to: restore-domain --domaindir /opt/dev/apps/domains/ --filename sjsas_backup_v00001.zip GiADomain The command will simply pick up the backup archive and restore the given domain using its content. Deleting a domain We should deal with this command very carefully as it will delete the domain directory with all of its content including backups, deployed applications, configuration files and so on. To delete a domain we can use delete-domain command as follow to delete our GiADomain domain. delete-domaind --domaindir=/opt/dev/apps/domains/ GiADomain Domain management is the most basic task related to application server administration. Every asadmin command including domain administration commands has manual pages included in the asadmin commands package and accessible using asadmin utility. An example of getting help for create-domain command is similar to: asadmin help create-domain You can see complete list of parameters and their associated values for each of the discussed command using the help command. All of the domain administration commands are local commands. We can use list-commands command to see the list of all available commands in both remote and local categories. 3.2 Applications lifecycle administration’ Applications including web, enterprise, and resource adapter, and so on are our bits inside the application server which our clients relay on to get the required services. Applications can be deployed from command line in a very effective and customizable manner. Like all other administration commands application deployment is a single parameter command when we want to use the default parameters. Deploying and un-deploying application To deploy an application independently from its type (WAR, EAR, RAR, and so on) we can use deploy command. The command which is one of the most basic remote commands deploys any kind of application known to the GlassFish application server to its dedicated container. The command has many options which are omitted to show the common use case. The GlassFish instance should be up and running to use deploy or any other remote command. deploy   --host  127.0.0.1 --port 4747  --user  admin –name aName --contextroot cRoot --upload=true --dbvendorname derby --createtables=true  /home/masoud/GiA.ear Deploying a web application using default parameters is as simple as passing the archive to the deploy command. The context name will be same as the archive file. deploy GiA.war We can remove an application using the undeploy command. The command will remove the application bits from the application server directories along with removing it from the list of known applications in the domain.xml file. Following snippet shows how we can use it. undeploy  --droptables=true aName The undeply command uses the application name to remove it from the application server; we have the liberty to drop the automatically created tables (for enterprise applications) automatically. Deploying and undeploying other application types is similar to enterprise application deployment. You can view the complete list of parameters by executing deploy with no parameter or you can view the complete help content by using help deploy command. Application utility commands For domains with developer profile we have only one utility command named list-applications which list the deployed applications. In addition to all shared parameter this command has one important parameter named --type which accepts application,    connector, ejb, jruby, and web as its value. Following command will list all web applications deployed in out domain. list-applications --type=web You can see all commands related to application administration by trying asadmin application which will show all commands which has application word in the command name. 3.3 Application server managed resources Dealing with managed resources is one of the most recurring tasks which an administrator faces during the application deployment and development. The most important types of managed resources are JDBC and JMS resources which we will discuss in this section. JDBC connection pool administration A JDBC connection pool is a group of reusable connections for a particular database. Because creating each new physical connection is time consuming, the server maintains a pool of available connections to increase performance. When an application requests a connection, it obtains one from the pool and when the application closes the connection, the connection is returned to the pool instead of being closed. To create a JDBC connection pool we can use create-jdbc-connection-pool command. create-jdbc-connection-pool --user admin --passwordfile=/opt/dev/apps/passfile.txt  --host localhost --port 4747 --datasourceclassname org.apache.derby.jdbc.ClientDataSource --restype javax.sql.XADataSource --property portNumber=1527:password=APP:user=APP:serverName= localhost:databaseName=GiADatabase:create=true GiA_Derby_Pool The command may looks scary at the beginning but if you take a closer look you can see that all of the parameters and values are well known for you. We ca delete a JDBC connection pool using delete-jdbc-connection-pool which in addition to shared parameters accept an operand which is the name of the connection pool which we want to delete Now that we have a connection pool we need a reference to the connection pool in the JNDI tree to ensure that our applications can access the connection pool through the application server. JDBC resources administration A JDBC resource or a data source is the mediator between application and the connection pool by providing a JNDI name to access the connection pool. Multiple JDBC resources can specify a single connection pool. create-jdbc-resource --user admin --port 4747 --host localhost --passwordfile=/opt/dev/apps/passfile.txt  --connectionpoolid GiA_Derby_Pool jdbc/GiAPool We determined a name for our JDBC connection pool when we create the pool. And when we want to create the data source we use that name to identity the JDBC connection pool we want to specify in the JDBC resource. We can delete a JDBC resource using delete-jdbc-resource which in addition to shared parameters accepts an operand which is name of the connection pool that we want to delete. JMS Destination administration In enterprise applications, asynchronous communication is one of the in-evitable requirements. JMS destinations are virtually places which each message are headed to. JMS clients are listening on the same destinations to receive these messages. We can create a JMS destination either a Queue or Topic by create-jmsdest command. create-jmsdest --user admin --passwordfile=/opt/dev/apps/passfile.txt --host localhost --port 4747 --desttype queue  --property User=public:Password=public GiAQueue We can create a JMS topic by changing the value of --desttype parameter to topic. To delete a JMS destination we can use delete-jmsdest command which accept the name of destination which we want to delete as an operand. For example the following command will delete the JMS queue which we create in previous step. delete-jmsdest GiAQueue Make sure that you create the above queue again as we will need it in the next steps. JMS resources administration JMS connection factories are door to consuming and producing JMS messages. We need to define these factories prior to be able to access a JMS destination. To create a JMS connection factory we can use creste-jms-resource command. For example: create-jms-resource --user admin --passwordfile=/opt/dev/apps/passfile.txt  --host localhost --port 4747  --restype javax.jms.QueueConnectionFactory --property ClientId=MyID:UserName=guest:Password=guest jms/GiAQueueConnectionFactory Our sample command creates a connection factory which provides us with queue connections. When we need to interact with a JMS topic we can change the --restype parameter to javax.jms.TopicConnectionFactory. Deleting a JMS resource is possible by using delete-jms-resource command which takes the resource name which we want to delete as a parameter in addition to the shared parameters. Utility commands There are several utility commands related to manage resources, table 3 shows these commands along with a sample and description.   Table 3 Some asadmin utility commands for administering managed resources Command Sample Description flush-jmsdest flush-jmsdest GiAQueue* purges the messages in a JMS destination jms-ping jms-ping checks to see if the JMS provider is running, the command can ping cluster wide configurations list-jms-resources ist-jms-resources --restype javax.jms.TopicConnectionFactory List all JMS resources list-jmsdest list-jmsdest List all JMS destinations list-jdbc-connection-pools list-jdbc-connection-pools List all JDBC connection pools list-jdbc-resources list-jdbc-resources List all JDBC data sources *I removed all shared parameters to fit the commands in the table. The asadmin utility has a very rich set of commands for administrating the managed resources. We learned basic commands here. 3 Monitoring with CLI Monitoring is an integral part of the administrators daily task list. Monitoring helps us to: find possible problems, detect performance bottlenecks, find system resource shortage, plan system capacity or upgrade, and finally monitor the daily performance of the system. The GlassFish application server infrastructure allows us to monitor almost any object which corresponds to a Java EE concept. For example we can monitor an HTTP listener, a Servlet, or a JDBC connection pool.  To monitor an object, system should gather statistics about that object and this statistics gathering has some overhead even in ideal software as more byte codes should be executed to gather the statistics. Based on the fact that GlassFish application server should be well tuned out of the box, all monitoring functionalities are turned off to prevent any extra overhead during the system work hours.  GlassFish provides 3 monitoring level for any object available for monitoring. These levels include:

• OFF: No monitoring information is gathered
• LOW: Only object’s attributes changes are monitored
• HIGH: In addition to attributes methods execution is counted

The asadmin utility provides multiple ways for getting statistical data from the application server core. Multiple commands are introduced to facilitate our job in viewing this statistical information. These commands include:

• The monitor command: We can use this command to view common statistical information about different application server services, we can filter the statistics for only on monitor-able object in the service or we can view cumulative statistical information about all objects in that particular service. For example all JDBC connection pools or just one single JDBC connection pool statistics can be viewed using the monitor command.
• The list command: We can use this command to get the hierarchic of all monitor-able and configuration objects or specific type of objects like JDBC connection pools. The list command has an important boolean parameter named monitor which determine whether we want to see monitor-able objects or configuration objects.
• The set command: We can use this command to change the monitoring level of a monitor-able object. And in broader view, We can use it to change application server configuration by changing the attributes of configuration objects.
• The get command: We can get customized monitoring information about different attributes of different monitor-able objects. In broader view, we can use this command to view value of any configuration object’s attributes.

3.1 Dotted names All monitor-able and configuration objects shape a tree composed of these objects and their children objects. The children objects have attributes which are either the monitoring factor or configuration elements. Objects in the tree are separated using a dot and therefore we can simply navigate from the root object to any of lower level child objects. For example if we execute list --monitor=true server we will get a list of all monitor-able objects which are direct child of server objects. The list can be similar to the following snippet. server.applications server.connector-service server.http-service server.jms-service server.jvm server.orb server.resources server.thread-pools server.transaction-service After getting the high level child we can go deeper toward leaf nodes by listing the immediate or related children of any top level object. For example to get immediate children or http-service object we can use list --monitor=true server.http-service to show all immediate children of http-service object which can result in an output similar to: server.http-service.__asadmin server.http-service.connection-queue server.http-service.file-cache server.http-service.keep-alive server.http-service.pwc-thread-pool server.http-service.server We can further do down and monitor any monitor-able object in the application server. We can use list --monitor=true server.http-service.* to view a recursive list of immediate children and children of those immediate children. We use * with list command as a place holder for any part of the dotted name. For example we can use server.jvm.* or server.resource* to get all children of server.jvm or all children which starts with resource. The dotted names are not provided for sole monitoring purposes, we can use them to view and change the configuration of application server by using the asadmin utility. We can get list of configurable objects by omitting the --monitor=true parameter from the list command. Enabling the GlassFish monitoring GlassFish monitoring is turned off when we install the application server, we need to enable the monitoring system to gather statistical information for us. To do so, we can use set command to change the monitoring level for one or all of the GlassFish services like JDBC connection pool, HTTP listener, and so on.  But before changing the monitoring level from OFF to HIGH we can check and see the current level by using a simple get command similar to the following snippet. get server.monitoring-service.module-monitoring-levels.* Executing this snippet shows us a result similar to the following snippet which means that monitoring is OFF for all services. server.monitoring-service.module-monitoring-levels.connector-service=OFF server.monitoring-service.module-monitoring-levels.ejb-container=OFF server.monitoring-service.module-monitoring-levels.http-service=OFF server.monitoring-service.module-monitoring-levels.jdbc-connection-pool=OFF server.monitoring-service.module-monitoring-levels.jms-service=OFF server.monitoring-service.module-monitoring-levels.jvm=OFF server.monitoring-service.module-monitoring-levels.orb=OFF server.monitoring-service.module-monitoring-levels.thread-pool=OFF server.monitoring-service.module-monitoring-levels.transaction-service=OFF server.monitoring-service.module-monitoring-levels.web-container=OFF We can enable the monitoring for JDBC connection pools by using set command as follow: Set server.monitoring-service.module-monitoring-levels.jdbc-connection-pool=HIGH Now we are sure that any activity involving any of the JDBC connection pool will be monitored and its statistical information will be available for us to view and analyze. 3.2 The monitor command Using monitor command is the simplest way to view monitoring information about GlassFish application server. Following snippet shows the monitor command syntax. I have not included the shared parameters to make it easier to read and see the command specific parameters. monitor --type  monitor_type [--filename file_name] [--interval interval] [--filter filter_name]   instance_name We can save the command output to CSV files using the --filename parameter and we can determine the interval which the command uses to retrieve the statistics by using the --interval parameter. The instance_name argument let us use this command on a DAS to retrieve the statistics related to one of the DAS member instances. Two other parameters determine what service and which object in the service we want to monitor. Most important acceptable values for --type parameter is listed in table 4 along with related description. Table 4 acceptable values for the monitor command’s type parameter Value Description statefulsession Statistics related to State full Session Beans statelesssession Statistics related to Stateless Session Beans connectorpool JCA connector pool statistics endpoint Web services end point statistics entitybean Entity Beans statistics filecache GlassFish HTTP level cache system statistics httplistener HTT listener statistics httpservice HTTP service statistics jdbcpool JDBC connection pool statistics jvm Underlying JVM statistics servlet Servlet statistics You can see an example monitoring command to monitor our newly created JDBC connection pool in the following snippet. monitor --type jdbcpool --interval 10 --filter GiA_Derby_Pool server We just want to view the statistics related to one JDBC connection pool on the default instance. The interval for retrieving the statistics is 10 seconds. The output for this command is very similar to figure 3. Figure 3 Outputs for monitoring GiA_Derby_Pool JDBC connection pool using monitor command Monitoring results are categorized under some column and for each column we have several statistic factors like highest number of free connections, and current number of free connections. Next service which we want to monitor is the application server runtime or the JVM instance that our application server is running on. To monitor the JVM we can use a command similar to the following snippet. monitor --type jvm --interval 1 server Command output is similar to figure 4 which shows almost no changes in heap size during the monitoring period. Figure 4 outputs for JVM monitoring using the monitor command The monitor command combined with LOW monitoring level is very useful when we need to get an overview of different GlassFish services in production environment. 3.3 Monitoring and configuration by get and set commands Explain more about get and set commands   Monitoring a JDBC connection pool Now we want to use these set of four commands to monitor our newly created JDBC connection pool.  get --monitor=true server.resources.GiA_Derby_Pool.* This command will return a handful set of statistical information about our connection pool. Although we can change the command in order to get only one of the connection pool attributes instead of complete statistical set. To get number of free connections in our JDBC pool we can use the following command get --monitor=true  server.resources.DerbyPool.numconnfree-current We can use get to view the configuration attributes of any configurable object in the application server. If we try to use get without --monitor=true, we should give try to view a configuration attribute instead of a monitoring attribute. For example to view the maximum pool size for our connection pool we can use the following command: get server.resources.jdbc-connection-pool.DerbyPool.max-pool-size We can use set command alongside with get command to change the configuration parameters using the dotted names and asadmin utility. set server.resources.jdbc-connection-pool.DerbyPool.max-pool-size=64 It is very simple and effective to get detailed monitoring information out of the application server to catch up with performance related issues. 4 CLI security The CLI is our means to administrate the application server. As well as it can be very useful for daily tasks it can be catastrophic if we fail to protect it well from the unauthorized access. 4.1 The asadmin related credentials In the commands which we studied in this article you saw that for some of them we used the --user and --passwordfile parameters and for some of them we did not. The main reason behind the fact that we could execute a command with or without providing any credentials refers back to the way that we create the domain. If you look at the command we used to create the domain you can see that we used --savemasterpassword=true and --savelogin=true.  These two parameters save the domain credentials and asadmin reuse those credentials whenever we want to execute a command. The so called master password The master password is designated to protect the domain encrypted files like digital certificate store from unauthorized access. When we start a GlassFish domain, startup process needs to read these certificates and therefore it need to open the certificates store files. GlassFish needs master password to open the store files and therefore we should either provide the master password during the start-domain command execution in an interactive way or we should use the --passwordfile parameter to provide the process with the master password. The most common situation for using a previously saved master password is when we need our application server to start in an unattended environment, like when we make a Windows service or Linux Daemon for it. To provide this password for the application server startup process, we should use the --savemasterpassword=true during the domain creation. If we use this parameter the master password will be saved into a file named master-password which resides inside the domain/config directory. We said that the master password protects some files from unauthorized access. One of the file set that master password protects is the digital certificates storage files which differ between developer and cluster profile. The domain administration credentials So far, we that we can execute asadmin commands with or without providing the password file. You may ask, how it is possible that we are not entering the password and nor we are providing the password file and hence we saw that our command are carried out by the  asadmin. We did not provide any password in an interactive manner or any password file for asadmin to carry out our command because we used a parameter during the domain creation. The --savelogin=true asks the domain creation command to save the administration password of our domain into a file name .asadminpass which is located inside our home directory. The content of this file is similar to the following snippet. asadmin://admin@localhost:4747 YWRtaW5hZG1pbg== The syntax simply determine what is the administration username and administration password for a domain that is running on localhost and uses 4747 as its administration port Sure the password is encrypted to prevent anyone learning it without being authorized. There is an asadmin command which perform a similar task to what --savelogin does for domains that are already created.  We can use this command to save the credentials to prevent the asadmin utility asking for them either in an interactive mode or as  --passwordfile parameter. Following snippet shows how we can use login command. login   --host localhost --port 9138 This command will interactively ask for administration username and administration password and after a successful authentication it will save the provided credentials into the .asadminpass file. After we execute this command the content of .asadminpass will be similar to the following snippet. asadmin://admin@localhost:4747 YWRtaW5hZG1pbg== asadmin://admin@localhost:9138 YWRtaW5hZG1pbg== The .asadminpass contains the SHA hashed copy of passwords therefore it is not possible for anyone to recover the original passwords if he can grasp the file. The login command falls very useful when we need to administrate several domains from our administration workstation. We can simply login into each remote or local domain that we need to administrate and then asadmin will pickup the correct credential from the .asadminpass file based on the --host and --port parameters.   Changing passwords As and administrator we usually like to change our passwords from time to time to ensure keep a higher level of security precautions. GlassFish let us simply change the master or administration password using some provided commands. To change the master password we must be sure that application server is not running and then we can the change-master-password as follow: Change-master-password --domaindir=/opt/dev/apps/domains/ --savemasterpassword=true GiADomain After executing this command, the asadmin utility will interactively asks us for the new master password which must be at least 8 characters. And we use the --savemasterpassword to ensure that the master password is saved and during the domain startup we do not need to provide the asadmin with it. We need to change the password file if we are using it to feed the asadmin utility with the master password. To change the administration password we can use change-admin-password which is a remote command and need the application server to be running. Following snippet shows how we can change the administration password for a given username. change-admin-password --host 127.0.0.1 --port 4747 --user admin After executing this command, asadmin will asks for administration password and then change the password of the given user. If we change the password for a user, we will need to login into that domain again if we need to use automatic login. Also we need to change the password file if we are using it to feed the asadmin utility with its required passwords. In advanced administration article we will discuss how we can add more administration users and how we can use an already credentials store like an LDAP for administration authentication. 4.2 Protecting passwords We have many places in any application server which we need to provide it wit some username and passwords which the application server will use to communicate with external systems like JMS brokers and Databases. Usually each part of the overall infrastructure has its own level of policy and permission set which lead to the fact that we should protect these passwords and avoid leaving these passwords in plain text format in any place, even in the application server configuration files which can be opened using any text editor. GlassFish provides a very elegant way for protecting these passwords by providing us with the required commands and infrastructure to encrypt the passwords and use the encrypted password’s assigned alias in the application server configuration files. The encrypted passwords are stored inside an encrypted file named domain-passwords which resides inside the domain’s config directory. The domain-passwords file is encrypted using the master password and if the master password compromised then these file can be decrypted. The command for creating password aliases is a remote command named create-password-alias and a sample usage is as following snippet: create-password-alias --user admin --host localhost --port 4747 GiA_Derby_Pool_Alias After we execute this command asadmin utility will ask for the password which we want this alias to hold.  Although asadmin may asks for administration credentials if we are not logged in. Now that we create the alias we can use it by using the alias accessing syntax which follows the ${ALIAS=password-alias-password} format. For example if we want to create the JDBC connection pool that we create in 3.3 we can change the command as follow: create-jdbc-connection-pool --user admin  --host localhost --port 4747 --datasourceclassname org.apache.derby.jdbc.ClientDataSource --restype javax.sql.XADataSource --property portNumber=1527:password=${ALIAS= GiA_Derby_Pool_Alias }:user=APP:serverName= localhost:databaseName=GiADatabase:create=true GiA_Derby_Pool   Password aliasing is not present just for external resources, but it can be used to protect the content of the password file which contains administration and master password for using instead of typing the password when the asadmin interactively asks for it.  We can simply create a password alias for administration password and for the master password and use them in password file. Sample content for a password file with aliased password is like. AS_ADMIN_PASSWORD=${ALIAS=admin-alias} AS_ADMIN_MAPPEDPASSWORD=${ALIAS=master-alias} Like all other administration commands, the alias administration commands set have some other commands which help with the commands administration. Other commands in the set are

• The delete-password-alias command: We can delete an alias when we are sure we are no longer using it.
• The list-password-aliases command: We can get a list of all aliases that we have in our domain-password file.
• The update-password-alias command: We can update an alias by changing the password that it holds.

Password aliasing is very helpful when we do not want to give our passwords to the personal in charge of application server management or administration tasks. Instead we can provide them aliased password which they can use. 5 Summary               The GlassFish command line administration interface is one of the most powerful and feature complete command line based administration utilities in the application server markets.  The asadmin utility provides all required commands which we need to administrate any part of the application server directly from the command line instead of wandering in the XML configuration files or third party utilities or the web based administration console. The asadmin utility is a not only a client side application which we can use to administrate an application and instead it is a composition of client side and a server side application which makes it possible to use the asadmin utility to administrate remote domains as well as local domains. The asadmin utility has enough intelligence embedded in its routines which helps us find the required commands and the command structure simply by entering a part of command. By using asadmin utility we can administrate domains, applications, and application server managed resources. When it comes to security, which is one of the highest concerns of the administrators, GlassFish provides many measures to protect passwords and keep the possibility of compromising passwords to the lowest possible value by using encryption for storing external resource passwords and asadmin administration passwords.       var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www."); document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E")); var pageTracker = _gat._getTracker("UA-137107-3"); pageTracker._initData(); pageTracker._trackPageview();

(I started cross-posting blogs to my own website.) As I wrote in my farewell note, I was working on starting a new company around Hudson. It took longer than I initially anticipated, but it's finally open for business! The company will provide two things; one is support, so that I can answer your questions and problem reports in a timely fashion, and the other is consulting, so that I can help you develop custom plugins, or provide on-site support to work on some tricky problems. The name of the company is InfraDNA because I think of Hudson more as an infrastructure on which all kinds of server-side automation/tools can be built/deployed, and because I think this stuff is built into me (as in DNA) — when I look back my career as a software engineer, I always somehow seem to come back to tooling. (Plus, the domain name was available!) Looking forward to hearing from you.