Search Results

Search found 3709 results on 149 pages for 'fusion middleware ibm ora'.

Page 149/149 | < Previous Page | 145 146 147 148 149 

  • Developing Spring Portlet for use inside Weblogic Portal / Webcenter Portal

    - by Murali Veligeti
    We need to understand the main difference between portlet workflow and servlet workflow.The main difference between portlet workflow and servlet workflow is that, the request to the portlet can have two distinct phases: 1) Action phase 2) Render phase. The Action phase is executed only once and is where any 'backend' changes or actions occur, such as making changes in a database. The Render phase then produces what is displayed to the user each time the display is refreshed. The critical point here is that for a single overall request, the action phase is executed only once, but the render phase may be executed multiple times. This provides a clean separation between the activities that modify the persistent state of your system and the activities that generate what is displayed to the user.The dual phases of portlet requests are one of the real strengths of the JSR-168 specification. For example, dynamic search results can be updated routinely on the display without the user explicitly re-running the search. Most other portlet MVC frameworks attempt to completely hide the two phases from the developer and make it look as much like traditional servlet development as possible - we think this approach removes one of the main benefits of using portlets. So, the separation of the two phases is preserved throughout the Spring Portlet MVC framework. The primary manifestation of this approach is that where the servlet version of the MVC classes will have one method that deals with the request, the portlet version of the MVC classes will have two methods that deal with the request: one for the action phase and one for the render phase. For example, where the servlet version of AbstractController has the handleRequestInternal(..) method, the portlet version of AbstractController has handleActionRequestInternal(..) and handleRenderRequestInternal(..) methods.The Spring Portlet Framework is designed around a DispatcherPortlet that dispatches requests to handlers, with configurable handler mappings and view resolution, just as the DispatcherServlet in the Spring Web Framework does.  Developing portlet.xml Let's start the sample development by creating the portlet.xml file in the /WebContent/WEB-INF/ folder as shown below: <?xml version="1.0" encoding="UTF-8"?> <portlet-app version="2.0" xmlns="http://java.sun.com/xml/ns/portlet/portlet-app_2_0.xsd" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"> <portlet> <portlet-name>SpringPortletName</portlet-name> <portlet-class>org.springframework.web.portlet.DispatcherPortlet</portlet-class> <supports> <mime-type>text/html</mime-type> <portlet-mode>view</portlet-mode> </supports> <portlet-info> <title>SpringPortlet</title> </portlet-info> </portlet> </portlet-app> DispatcherPortlet is responsible for handling every client request. When it receives a request, it finds out which Controller class should be used for handling this request, and then it calls its handleActionRequest() or handleRenderRequest() method based on the request processing phase. The Controller class executes business logic and returns a View name that should be used for rendering markup to the user. The DispatcherPortlet then forwards control to that View for actual markup generation. As you can see, DispatcherPortlet is the central dispatcher for use within Spring Portlet MVC Framework. Note that your portlet application can define more than one DispatcherPortlet. If it does so, then each of these portlets operates its own namespace, loading its application context and handler mapping. The DispatcherPortlet is also responsible for loading application context (Spring configuration file) for this portlet. First, it tries to check the value of the configLocation portlet initialization parameter. If that parameter is not specified, it takes the portlet name (that is, the value of the <portlet-name> element), appends "-portlet.xml" to it, and tries to load that file from the /WEB-INF folder. In the portlet.xml file, we did not specify the configLocation initialization parameter, so let's create SpringPortletName-portlet.xml file in the next section. Developing SpringPortletName-portlet.xml Create the SpringPortletName-portlet.xml file in the /WebContent/WEB-INF folder of your application as shown 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" xsi:schemaLocation="http://www.springframework.org/schema/beans http://www.springframework.org/schema/beans/spring-beans-2.0.xsd"> <bean id="viewResolver" class="org.springframework.web.servlet.view.InternalResourceViewResolver"> <property name="viewClass" value="org.springframework.web.servlet.view.JstlView"/> <property name="prefix" value="/jsp/"/> <property name="suffix" value=".jsp"/> </bean> <bean id="pointManager" class="com.wlp.spring.bo.internal.PointManagerImpl"> <property name="users"> <list> <ref bean="point1"/> <ref bean="point2"/> <ref bean="point3"/> <ref bean="point4"/> </list> </property> </bean> <bean id="point1" class="com.wlp.spring.bean.User"> <property name="name" value="Murali"/> <property name="points" value="6"/> </bean> <bean id="point2" class="com.wlp.spring.bean.User"> <property name="name" value="Sai"/> <property name="points" value="13"/> </bean> <bean id="point3" class="com.wlp.spring.bean.User"> <property name="name" value="Rama"/> <property name="points" value="43"/> </bean> <bean id="point4" class="com.wlp.spring.bean.User"> <property name="name" value="Krishna"/> <property name="points" value="23"/> </bean> <bean id="messageSource" class="org.springframework.context.support.ResourceBundleMessageSource"> <property name="basename" value="messages"/> </bean> <bean name="/users.htm" id="userController" class="com.wlp.spring.controller.UserController"> <property name="pointManager" ref="pointManager"/> </bean> <bean name="/pointincrease.htm" id="pointIncreaseController" class="com.wlp.spring.controller.IncreasePointsFormController"> <property name="sessionForm" value="true"/> <property name="pointManager" ref="pointManager"/> <property name="commandName" value="pointIncrease"/> <property name="commandClass" value="com.wlp.spring.bean.PointIncrease"/> <property name="formView" value="pointincrease"/> <property name="successView" value="users"/> </bean> <bean id="parameterMappingInterceptor" class="org.springframework.web.portlet.handler.ParameterMappingInterceptor" /> <bean id="portletModeParameterHandlerMapping" class="org.springframework.web.portlet.handler.PortletModeParameterHandlerMapping"> <property name="order" value="1" /> <property name="interceptors"> <list> <ref bean="parameterMappingInterceptor" /> </list> </property> <property name="portletModeParameterMap"> <map> <entry key="view"> <map> <entry key="pointincrease"> <ref bean="pointIncreaseController" /> </entry> <entry key="users"> <ref bean="userController" /> </entry> </map> </entry> </map> </property> </bean> <bean id="portletModeHandlerMapping" class="org.springframework.web.portlet.handler.PortletModeHandlerMapping"> <property name="order" value="2" /> <property name="portletModeMap"> <map> <entry key="view"> <ref bean="userController" /> </entry> </map> </property> </bean> </beans> The SpringPortletName-portlet.xml file is an application context file for your MVC portlet. It has a couple of bean definitions: viewController. At this point, remember that the viewController bean definition points to the com.ibm.developerworks.springmvc.ViewController.java class. portletModeHandlerMapping. As we discussed in the last section, whenever DispatcherPortlet gets a client request, it tries to find a suitable Controller class for handling that request. That is where PortletModeHandlerMapping comes into the picture. The PortletModeHandlerMapping class is a simple implementation of the HandlerMapping interface and is used by DispatcherPortlet to find a suitable Controller for every request. The PortletModeHandlerMapping class uses Portlet mode for the current request to find a suitable Controller class to use for handling the request. The portletModeMap property of portletModeHandlerMapping bean is the place where we map the Portlet mode name against the Controller class. In the sample code, we show that viewController is responsible for handling View mode requests. Developing UserController.java In the preceding section, you learned that the viewController bean is responsible for handling all the View mode requests. Your next step is to create the UserController.java class as shown below: public class UserController extends AbstractController { private PointManager pointManager; public void handleActionRequest(ActionRequest request, ActionResponse response) throws Exception { } public ModelAndView handleRenderRequest(RenderRequest request, RenderResponse response) throws ServletException, IOException { String now = (new java.util.Date()).toString(); Map<String, Object> myModel = new HashMap<String, Object>(); myModel.put("now", now); myModel.put("users", this.pointManager.getUsers()); return new ModelAndView("users", "model", myModel); } public void setPointManager(PointManager pointManager) { this.pointManager = pointManager; } } Every controller class in Spring Portlet MVC Framework must implement the org.springframework.web. portlet.mvc.Controller interface directly or indirectly. To make things easier, Spring Framework provides AbstractController class, which is the default implementation of the Controller interface. As a developer, you should always extend your controller from either AbstractController or one of its more specific subclasses. Any implementation of the Controller class should be reusable, thread-safe, and capable of handling multiple requests throughout the lifecycle of the portlet. In the sample code, we create the ViewController class by extending it from AbstractController. Because we don't want to do any action processing in the HelloSpringPortletMVC portlet, we override only the handleRenderRequest() method of AbstractController. Now, the only thing that HelloWorldPortletMVC should do is render the markup of View.jsp to the user when it receives a user request to do so. To do that, return the object of ModelAndView with a value of view equal to View. Developing web.xml According to Portlet Specification 1.0, every portlet application is also a Servlet Specification 2.3-compliant Web application, and it needs a Web application deployment descriptor (that is, web.xml). Let’s create the web.xml file in the /WEB-INF/ folder as shown in listing 4. Follow these steps: Open the existing web.xml file located at /WebContent/WEB-INF/web.xml. Replace the contents of this file with the code as shown below: <servlet> <servlet-name>ViewRendererServlet</servlet-name> <servlet-class>org.springframework.web.servlet.ViewRendererServlet</servlet-class> </servlet> <servlet-mapping> <servlet-name>ViewRendererServlet</servlet-name> <url-pattern>/WEB-INF/servlet/view</url-pattern> </servlet-mapping> <context-param> <param-name>contextConfigLocation</param-name> <param-value>/WEB-INF/applicationContext.xml</param-value> </context-param> <listener> <listener-class>org.springframework.web.context.ContextLoaderListener</listener-class> </listener> The web.xml file for the sample portlet declares two things: ViewRendererServlet. The ViewRendererServlet is the bridge servlet for portlet support. During the render phase, DispatcherPortlet wraps PortletRequest into ServletRequest and forwards control to ViewRendererServlet for actual rendering. This process allows Spring Portlet MVC Framework to use the same View infrastructure as that of its servlet version, that is, Spring Web MVC Framework. ContextLoaderListener. The ContextLoaderListener class takes care of loading Web application context at the time of the Web application startup. The Web application context is shared by all the portlets in the portlet application. In case of duplicate bean definition, the bean definition in the portlet application context takes precedence over the Web application context. The ContextLoader class tries to read the value of the contextConfigLocation Web context parameter to find out the location of the context file. If the contextConfigLocation parameter is not set, then it uses the default value, which is /WEB-INF/applicationContext.xml, to load the context file. The Portlet Controller interface requires two methods that handle the two phases of a portlet request: the action request and the render request. The action phase should be capable of handling an action request and the render phase should be capable of handling a render request and returning an appropriate model and view. While the Controller interface is quite abstract, Spring Portlet MVC offers a lot of controllers that already contain a lot of the functionality you might need – most of these are very similar to controllers from Spring Web MVC. The Controller interface just defines the most common functionality required of every controller - handling an action request, handling a render request, and returning a model and a view. How rendering works As you know, when the user tries to access a page with PointSystemPortletMVC portlet on it or when the user performs some action on any other portlet on that page or tries to refresh that page, a render request is sent to the PointSystemPortletMVC portlet. In the sample code, because DispatcherPortlet is the main portlet class, Weblogic Portal / Webcenter Portal calls its render() method and then the following sequence of events occurs: The render() method of DispatcherPortlet calls the doDispatch() method, which in turn calls the doRender() method. After the doRenderService() method gets control, first it tries to find out the locale of the request by calling the PortletRequest.getLocale() method. This locale is used while making all the locale-related decisions for choices such as which resource bundle should be loaded or which JSP should be displayed to the user based on the locale. After that, the doRenderService() method starts iterating through all the HandlerMapping classes configured for this portlet, calling their getHandler() method to identify the appropriate Controller for handling this request. In the sample code, we have configured only PortletModeHandlerMapping as a HandlerMapping class. The PortletModeHandlerMapping class reads the value of the current portlet mode, and based on that, it finds out, the Controller class that should be used to handle this request. In the sample code, ViewController is configured to handle the View mode request so that the PortletModeHandlerMapping class returns the object of ViewController. After the object of ViewController is returned, the doRenderService() method calls its handleRenderRequestInternal() method. Implementation of the handleRenderRequestInternal() method in ViewController.java is very simple. It logs a message saying that it got control, and then it creates an instance of ModelAndView with a value equal to View and returns it to DispatcherPortlet. After control returns to doRenderService(), the next task is to figure out how to render View. For that, DispatcherPortlet starts iterating through all the ViewResolvers configured in your portlet application, calling their resolveViewName() method. In the sample code we have configured only one ViewResolver, InternalResourceViewResolver. When its resolveViewName() method is called with viewName, it tries to add /WEB-INF/jsp as a prefix to the view name and to add JSP as a suffix. And it checks if /WEB-INF/jsp/View.jsp exists. If it does exist, it returns the object of JstlView wrapping View.jsp. After control is returned to the doRenderService() method, it creates the object PortletRequestDispatcher, which points to /WEB-INF/servlet/view – that is, ViewRendererServlet. Then it sets the object of JstlView in the request and dispatches the request to ViewRendererServlet. After ViewRendererServlet gets control, it reads the JstlView object from the request attribute and creates another RequestDispatcher pointing to the /WEB-INF/jsp/View.jsp URL and passes control to it for actual markup generation. The markup generated by View.jsp is returned to user. At this point, you may question the need for ViewRendererServlet. Why can't DispatcherPortlet directly forward control to View.jsp? Adding ViewRendererServlet in between allows Spring Portlet MVC Framework to reuse the existing View infrastructure. You may appreciate this more when we discuss how easy it is to integrate Apache Tiles Framework with your Spring Portlet MVC Framework. The attached project SpringPortlet.zip should be used to import the project in to your OEPE Workspace. SpringPortlet_Jars.zip contains jar files required for the application. Project is written on Spring 2.5.  The same JSR 168 portlet should work on Webcenter Portal as well.  Downloads: Download WeblogicPotal Project which consists of Spring Portlet. Download Spring Jars In-addition to above you need to download Spring.jar (Spring2.5)

    Read the article

  • JMS Step 5 - How to Create an 11g BPEL Process Which Reads a Message Based on an XML Schema from a JMS Queue

    - by John-Brown.Evans
    JMS Step 5 - How to Create an 11g BPEL Process Which Reads a Message Based on an XML Schema from a JMS Queue .jblist{list-style-type:disc;margin:0;padding:0;padding-left:0pt;margin-left:36pt} ol{margin:0;padding:0} .c12_5{vertical-align:top;width:468pt;border-style:solid;background-color:#f3f3f3;border-color:#000000;border-width:1pt;padding:5pt 5pt 5pt 5pt} .c8_5{vertical-align:top;border-style:solid;border-color:#000000;border-width:1pt;padding:5pt 5pt 0pt 5pt} .c10_5{vertical-align:top;width:207pt;border-style:solid;border-color:#000000;border-width:1pt;padding:5pt 5pt 5pt 5pt} .c14_5{vertical-align:top;border-style:solid;border-color:#000000;border-width:1pt;padding:0pt 5pt 0pt 5pt} .c21_5{background-color:#ffffff} .c18_5{color:#1155cc;text-decoration:underline} .c16_5{color:#666666;font-size:12pt} .c5_5{background-color:#f3f3f3;font-weight:bold} .c19_5{color:inherit;text-decoration:inherit} .c3_5{height:11pt;text-align:center} .c11_5{font-weight:bold} .c20_5{background-color:#00ff00} .c6_5{font-style:italic} .c4_5{height:11pt} .c17_5{background-color:#ffff00} .c0_5{direction:ltr} .c7_5{font-family:"Courier New"} .c2_5{border-collapse:collapse} .c1_5{line-height:1.0} .c13_5{background-color:#f3f3f3} .c15_5{height:0pt} .c9_5{text-align:center} .title{padding-top:24pt;line-height:1.15;text-align:left;color:#000000;font-size:36pt;font-family:"Arial";font-weight:bold;padding-bottom:6pt} .subtitle{padding-top:18pt;line-height:1.15;text-align:left;color:#666666;font-style:italic;font-size:24pt;font-family:"Georgia";padding-bottom:4pt} li{color:#000000;font-size:10pt;font-family:"Arial"} p{color:#000000;font-size:10pt;margin:0;font-family:"Arial"} h1{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:24pt;font-family:"Arial";font-weight:normal} h2{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:18pt;font-family:"Arial";font-weight:normal} h3{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:14pt;font-family:"Arial";font-weight:normal} h4{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:12pt;font-family:"Arial";font-weight:normal} h5{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:11pt;font-family:"Arial";font-weight:normal} h6{padding-top:0pt;line-height:1.15;text-align:left;color:#888;font-size:10pt;font-family:"Arial";font-weight:normal} Welcome to another post in the series of blogs which demonstrates how to use JMS queues in a SOA context. The previous posts were: JMS Step 1 - How to Create a Simple JMS Queue in Weblogic Server 11g JMS Step 2 - Using the QueueSend.java Sample Program to Send a Message to a JMS Queue JMS Step 3 - Using the QueueReceive.java Sample Program to Read a Message from a JMS Queue JMS Step 4 - How to Create an 11g BPEL Process Which Writes a Message Based on an XML Schema to a JMS Queue Today we will create a BPEL process which will read (dequeue) the message from the JMS queue, which we enqueued in the last example. The JMS adapter will dequeue the full XML payload from the queue. 1. Recap and Prerequisites In the previous examples, we created a JMS Queue, a Connection Factory and a Connection Pool in the WebLogic Server Console. Then we designed and deployed a BPEL composite, which took a simple XML payload and enqueued it to the JMS queue. In this example, we will read that same message from the queue, using a JMS adapter and a BPEL process. As many of the configuration steps required to read from that queue were done in the previous samples, this one will concentrate on the new steps. A summary of the required objects is listed below. To find out how to create them please see the previous samples. They also include instructions on how to verify the objects are set up correctly. WebLogic Server Objects Object Name Type JNDI Name TestConnectionFactory Connection Factory jms/TestConnectionFactory TestJMSQueue JMS Queue jms/TestJMSQueue eis/wls/TestQueue Connection Pool eis/wls/TestQueue Schema XSD File The following XSD file is used for the message format. It was created in the previous example and will be copied to the new process. stringPayload.xsd <?xml version="1.0" encoding="windows-1252" ?> <xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema"                 xmlns="http://www.example.org"                 targetNamespace="http://www.example.org"                 elementFormDefault="qualified">   <xsd:element name="exampleElement" type="xsd:string">   </xsd:element> </xsd:schema> JMS Message After executing the previous samples, the following XML message should be in the JMS queue located at jms/TestJMSQueue: <?xml version="1.0" encoding="UTF-8" ?><exampleElement xmlns="http://www.example.org">Test Message</exampleElement> JDeveloper Connection You will need a valid Application Server Connection in JDeveloper pointing to the SOA server which the process will be deployed to. 2. Create a BPEL Composite with a JMS Adapter Partner Link In the previous example, we created a composite in JDeveloper called JmsAdapterWriteSchema. In this one, we will create a new composite called JmsAdapterReadSchema. There are probably many ways of incorporating a JMS adapter into a SOA composite for incoming messages. One way is design the process in such a way that the adapter polls for new messages and when it dequeues one, initiates a SOA or BPEL instance. This is possibly the most common use case. Other use cases include mid-flow adapters, which are activated from within the BPEL process. In this example we will use a polling adapter, because it is the most simple to set up and demonstrate. But it has one disadvantage as a demonstrative model. When a polling adapter is active, it will dequeue all messages as soon as they reach the queue. This makes it difficult to monitor messages we are writing to the queue, because they will disappear from the queue as soon as they have been enqueued. To work around this, we will shut down the composite after deploying it and restart it as required. (Another solution for this would be to pause the consumption for the queue and resume consumption again if needed. This can be done in the WLS console JMS-Modules -> queue -> Control -> Consumption -> Pause/Resume.) We will model the composite as a one-way incoming process. Usually, a BPEL process will do something useful with the message after receiving it, such as passing it to a database or file adapter, a human workflow or external web service. But we only want to demonstrate how to dequeue a JMS message using BPEL and a JMS adapter, so we won’t complicate the design with further activities. However, we do want to be able to verify that we have read the message correctly, so the BPEL process will include a small piece of embedded java code, which will print the message to standard output, so we can view it in the SOA server’s log file. Alternatively, you can view the instance in the Enterprise Manager and verify the message. The following steps are all executed in JDeveloper. Create the project in the same JDeveloper application used for the previous examples or create a new one. Create a SOA Project Create a new project and choose SOA Tier > SOA Project as its type. Name it JmsAdapterReadSchema. When prompted for the composite type, choose Empty Composite. Create a JMS Adapter Partner Link In the composite editor, drag a JMS adapter over from the Component Palette to the left-hand swim lane, under Exposed Services. This will start the JMS Adapter Configuration Wizard. Use the following entries: Service Name: JmsAdapterRead Oracle Enterprise Messaging Service (OEMS): Oracle WebLogic JMS AppServer Connection: Use an application server connection pointing to the WebLogic server on which the JMS queue and connection factory mentioned under Prerequisites above are located. Adapter Interface > Interface: Define from operation and schema (specified later) Operation Type: Consume Message Operation Name: Consume_message Consume Operation Parameters Destination Name: Press the Browse button, select Destination Type: Queues, then press Search. Wait for the list to populate, then select the entry for TestJMSQueue , which is the queue created in a previous example. JNDI Name: The JNDI name to use for the JMS connection. As in the previous example, this is probably the most common source of error. This is the JNDI name of the JMS adapter’s connection pool created in the WebLogic Server and which points to the connection factory. JDeveloper does not verify the value entered here. If you enter a wrong value, the JMS adapter won’t find the queue and you will get an error message at runtime, which is very difficult to trace. In our example, this is the value eis/wls/TestQueue . (See the earlier step on how to create a JMS Adapter Connection Pool in WebLogic Server for details.) Messages/Message SchemaURL: We will use the XSD file created during the previous example, in the JmsAdapterWriteSchema project to define the format for the incoming message payload and, at the same time, demonstrate how to import an existing XSD file into a JDeveloper project. Press the magnifying glass icon to search for schema files. In the Type Chooser, press the Import Schema File button. Select the magnifying glass next to URL to search for schema files. Navigate to the location of the JmsAdapterWriteSchema project > xsd and select the stringPayload.xsd file. Check the “Copy to Project” checkbox, press OK and confirm the following Localize Files popup. Now that the XSD file has been copied to the local project, it can be selected from the project’s schema files. Expand Project Schema Files > stringPayload.xsd and select exampleElement: string . Press Next and Finish, which will complete the JMS Adapter configuration.Save the project. Create a BPEL Component Drag a BPEL Process from the Component Palette (Service Components) to the Components section of the composite designer. Name it JmsAdapterReadSchema and select Template: Define Service Later and press OK. Wire the JMS Adapter to the BPEL Component Now wire the JMS adapter to the BPEL process, by dragging the arrow from the adapter to the BPEL process. A Transaction Properties popup will be displayed. Set the delivery mode to async.persist. This completes the steps at the composite level. 3 . Complete the BPEL Process Design Invoke the BPEL Flow via the JMS Adapter Open the BPEL component by double-clicking it in the design view of the composite.xml, or open it from the project navigator by selecting the JmsAdapterReadSchema.bpel file. This will display the BPEL process in the design view. You should see the JmsAdapterRead partner link in the left-hand swim lane. Drag a Receive activity onto the BPEL flow diagram, then drag a wire (left-hand yellow arrow) from it to the JMS adapter. This will open the Receive activity editor. Auto-generate the variable by pressing the green “+” button and check the “Create Instance” checkbox. This will result in a BPEL instance being created when a new JMS message is received. At this point it would actually be OK to compile and deploy the composite and it would pick up any messages from the JMS queue. In fact, you can do that to test it, if you like. But it is very rudimentary and would not be doing anything useful with the message. Also, you could only verify the actual message payload by looking at the instance’s flow in the Enterprise Manager. There are various other possibilities; we could pass the message to another web service, write it to a file using a file adapter or to a database via a database adapter etc. But these will all introduce unnecessary complications to our sample. So, to keep it simple, we will add a small piece of Java code to the BPEL process which will write the payload to standard output. This will be written to the server’s log file, which will be easy to monitor. Add a Java Embedding Activity First get the full name of the process’s input variable, as this will be needed for the Java code. Go to the Structure pane and expand Variables > Process > Variables. Then expand the input variable, for example, "Receive1_Consume_Message_InputVariable > body > ns2:exampleElement”, and note variable’s name and path, if they are different from this one. Drag a Java Embedding activity from the Component Palette (Oracle Extensions) to the BPEL flow, after the Receive activity, then open it to edit. Delete the example code and replace it with the following, replacing the variable parts with those in your sample, if necessary.: System.out.println("JmsAdapterReadSchema process picked up a message"); oracle.xml.parser.v2.XMLElement inputPayload =    (oracle.xml.parser.v2.XMLElement)getVariableData(                           "Receive1_Consume_Message_InputVariable",                           "body",                           "/ns2:exampleElement");   String inputString = inputPayload.getFirstChild().getNodeValue(); System.out.println("Input String is " + inputPayload.getFirstChild().getNodeValue()); Tip. If you are not sure of the exact syntax of the input variable, create an Assign activity in the BPEL process and copy the variable to another, temporary one. Then check the syntax created by the BPEL designer. This completes the BPEL process design in JDeveloper. Save, compile and deploy the process to the SOA server. 3. Test the Composite Shut Down the JmsAdapterReadSchema Composite After deploying the JmsAdapterReadSchema composite to the SOA server it is automatically activated. If there are already any messages in the queue, the adapter will begin polling them. To ease the testing process, we will deactivate the process first Log in to the Enterprise Manager (Fusion Middleware Control) and navigate to SOA > soa-infra (soa_server1) > default (or wherever you deployed your composite to) and click on JmsAdapterReadSchema [1.0] . Press the Shut Down button to disable the composite and confirm the following popup. Monitor Messages in the JMS Queue In a separate browser window, log in to the WebLogic Server Console and navigate to Services > Messaging > JMS Modules > TestJMSModule > TestJMSQueue > Monitoring. This is the location of the JMS queue we created in an earlier sample (see the prerequisites section of this sample). Check whether there are any messages already in the queue. If so, you can dequeue them using the QueueReceive Java program created in an earlier sample. This will ensure that the queue is empty and doesn’t contain any messages in the wrong format, which would cause the JmsAdapterReadSchema to fail. Send a Test Message In the Enterprise Manager, navigate to the JmsAdapterWriteSchema created earlier, press Test and send a test message, for example “Message from JmsAdapterWriteSchema”. Confirm that the message was written correctly to the queue by verifying it via the queue monitor in the WLS Console. Monitor the SOA Server’s Output A program deployed on the SOA server will write its standard output to the terminal window in which the server was started, unless this has been redirected to somewhere else, for example to a file. If it has not been redirected, go to the terminal session in which the server was started, otherwise open and monitor the file to which it was redirected. Re-Enable the JmsAdapterReadSchema Composite In the Enterprise Manager, navigate to the JmsAdapterReadSchema composite again and press Start Up to re-enable it. This should cause the JMS adapter to dequeue the test message and the following output should be written to the server’s standard output: JmsAdapterReadSchema process picked up a message. Input String is Message from JmsAdapterWriteSchema Note that you can also monitor the payload received by the process, by navigating to the the JmsAdapterReadSchema’s Instances tab in the Enterprise Manager. Then select the latest instance and view the flow of the BPEL component. The Receive activity will contain and display the dequeued message too. 4 . Troubleshooting This sample demonstrates how to dequeue an XML JMS message using a BPEL process and no additional functionality. For example, it doesn’t contain any error handling. Therefore, any errors in the payload will result in exceptions being written to the log file or standard output. If you get any errors related to the payload, such as Message handle error ... ORABPEL-09500 ... XPath expression failed to execute. An error occurs while processing the XPath expression; the expression is /ns2:exampleElement. ... etc. check that the variable used in the Java embedding part of the process was entered correctly. Possibly follow the tip mentioned in previous section. If this doesn’t help, you can delete the Java embedding part and simply verify the message via the flow diagram in the Enterprise Manager. Or use a different method, such as writing it to a file via a file adapter. This concludes this example. In the next post, we will begin with an AQ JMS example, which uses JMS to write to an Advanced Queue stored in the database. Best regards John-Brown Evans Oracle Technology Proactive Support Delivery

    Read the article

  • Where is the sample applications in the lastest Spring release(Spring Framework 3.0.2)?

    - by Yousui
    Hi guys, On the Spring download page, It says that For all Spring Framework releases, the basic release contains only the binaries while the -with-dependencies release contains everything the basic release contains plus all third-party dependencies, buildable source trees, and sample applications. When I download the spring-framework-3.0.2.RELEASE-dependencies.zip, after extract it I get a list of folders: I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.bea.commonj I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.caucho I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.google.jarjar I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.h2database I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.ibm.websphere I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.jamonapi I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.lowagie.text I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.mchange.c3p0 I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.opensymphony.quartz I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.oracle.toplink.essentials I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.springsource.bundlor I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.springsource.util I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.sun.msv I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.sun.syndication I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.sun.xml I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\com.thoughtworks.xstream I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\edu.emory.mathcs.backport I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\edu.oswego.cs.concurrent I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.activation I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.annotation I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.ejb I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.el I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.faces I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.inject I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.jdo I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.jms I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.mail I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.persistence I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.portlet I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.resource I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.servlet I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.transaction I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.validation I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.xml.bind I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.xml.rpc I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.xml.soap I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.xml.stream I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\javax.xml.ws I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\net.sourceforge.cglib I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\net.sourceforge.ehcache I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\net.sourceforge.iso-relax I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\net.sourceforge.jasperreports I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\net.sourceforge.jexcelapi I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\net.sourceforge.jibx I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\net.sourceforge.serp I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\net.sourceforge.xslthl I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.antlr I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.aopalliance I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.axis I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.bcel I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.catalina I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.commons I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.coyote I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.derby I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.ibatis I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.juli I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.log4j I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.openjpa I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.poi I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.regexp I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.struts I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.taglibs I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.tiles I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.velocity I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.xerces I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.xml I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.xmlbeans I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.apache.xmlcommons I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.aspectj I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.beanshell I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.codehaus.castor I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.codehaus.groovy I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.codehaus.jackson I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.codehaus.jettison I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.codehaus.woodstox I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.custommonkey.xmlunit I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.dom4j I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.easymock I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.eclipse.jdt I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.eclipse.persistence I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.freemarker I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.hibernate I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.hsqldb I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.jaxen I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.jboss.javassist I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.jboss.logging I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.jboss.util I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.jboss.vfs I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.jdom I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.jgroups I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.joda I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.jruby I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.junit I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.jvnet.staxex I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.mortbay.jetty I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.mozilla.javascript I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.objectweb.asm I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.osgi I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.relaxng I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.slf4j I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.springframework I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.springframework.build I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.testng I:\soft\java\spring-framework-3.0.2.RELEASE-dependencies\org.xmlpull So where are the sample applications? I know one of the sample applications is called jpetstore in spring 2.0. I did search in these folders and can't find anything useful. By the way, I also download the basic release which is spring-framework-3.0.2.RELEASE.zip. In the readme.txt of the basic release I found the following text: GETTING STARTED Please consult the blog examples at http://blog.springsource.com as well as the sections of interest in the reference documentation. Sample applications and related material will be provided as separate downloads. But I still don't know where to download the sample applications. Anyone can help? Thanks in advance.

    Read the article

  • Bacula & Multiple Tape Devices, and so on

    - by Tom O'Connor
    Bacula won't make use of 2 tape devices simultaneously. (Search for #-#-# for the TL;DR) A little background, perhaps. In the process of trying to get a decent working backup solution (backing up 20TB ain't cheap, or easy) at $dayjob, we bought a bunch of things to make it work. Firstly, there's a Spectra Logic T50e autochanger, 40 slots of LTO5 goodness, and that robot's got a pair of IBM HH5 Ultrium LTO5 drives, connected via FibreChannel Arbitrated Loop to our backup server. There's the backup server.. A Dell R715 with 2x 16 core AMD 62xx CPUs, and 32GB of RAM. Yummy. That server's got 2 Emulex FCe-12000E cards, and an Intel X520-SR dual port 10GE NIC. We were also sold Commvault Backup (non-NDMP). Here's where it gets really complicated. Spectra Logic and Commvault both sent respective engineers, who set up the library and the software. Commvault was running fine, in so far as the controller was working fine. The Dell server has Ubuntu 12.04 server, and runs the MediaAgent for CommVault, and mounts our BlueArc NAS as NFS to a few mountpoints, like /home, and some stuff in /mnt. When backing up from the NFS mountpoints, we were seeing ~= 290GB/hr throughput. That's CRAP, considering we've got 20-odd TB to get through, in a <48 hour backup window. The rated maximum on the BlueArc is 700MB/s (2460GB/hr), the rated maximum write speed on the tape devices is 140MB/s, per drive, so that's 492GB/hr (or double it, for the total throughput). So, the next step was to benchmark NFS performance with IOzone, and it turns out that we get epic write performance (across 20 threads), and it's like 1.5-2.5TB/hr write, but read performance is fecking hopeless. I couldn't ever get higher than 343GB/hr maximum. So let's assume that the 343GB/hr is a theoretical maximum for read performance on the NAS, then we should in theory be able to get that performance out of a) CommVault, and b) any other backup agent. Not the case. Commvault seems to only ever give me 200-250GB/hr throughput, and out of experimentation, I installed Bacula to see what the state of play there is. If, for example, Bacula gave consistently better performance and speeds than Commvault, then we'd be able to say "**$.$ Refunds Plz $.$**" #-#-# Alas, I found a different problem with Bacula. Commvault seems pretty happy to read from one part of the mountpoint with one thread, and stream that to a Tape device, whilst reading from some other directory with the other thread, and writing to the 2nd drive in the autochanger. I can't for the life of me get Bacula to mount and write to two tape drives simultaneously. Things I've tried: Setting Maximum Concurrent Jobs = 20 in the Director, File and Storage Daemons Setting Prefer Mounted Volumes = no in the Job Definition Setting multiple devices in the Autochanger resource. Documentation seems to be very single-drive centric, and we feel a little like we've strapped a rocket to a hamster, with this one. The majority of example Bacula configurations are for DDS4 drives, manual tape swapping, and FreeBSD or IRIX systems. I should probably add that I'm not too bothered if this isn't possible, but I'd be surprised. I basically want to use Bacula as proof to stick it to the software vendors that they're overpriced ;) I read somewhere that @KyleBrandt has done something similar with a modern Tape solution.. Configuration Files: *bacula-dir.conf* # # Default Bacula Director Configuration file Director { # define myself Name = backuphost-1-dir DIRport = 9101 # where we listen for UA connections QueryFile = "/etc/bacula/scripts/query.sql" WorkingDirectory = "/var/lib/bacula" PidDirectory = "/var/run/bacula" Maximum Concurrent Jobs = 20 Password = "yourekiddingright" # Console password Messages = Daemon DirAddress = 0.0.0.0 #DirAddress = 127.0.0.1 } JobDefs { Name = "DefaultFileJob" Type = Backup Level = Incremental Client = backuphost-1-fd FileSet = "Full Set" Schedule = "WeeklyCycle" Storage = File Messages = Standard Pool = File Priority = 10 Write Bootstrap = "/var/lib/bacula/%c.bsr" } JobDefs { Name = "DefaultTapeJob" Type = Backup Level = Incremental Client = backuphost-1-fd FileSet = "Full Set" Schedule = "WeeklyCycle" Storage = "SpectraLogic" Messages = Standard Pool = AllTapes Priority = 10 Write Bootstrap = "/var/lib/bacula/%c.bsr" Prefer Mounted Volumes = no } # # Define the main nightly save backup job # By default, this job will back up to disk in /nonexistant/path/to/file/archive/dir Job { Name = "BackupClient1" JobDefs = "DefaultFileJob" } Job { Name = "BackupThisVolume" JobDefs = "DefaultTapeJob" FileSet = "SpecialVolume" } #Job { # Name = "BackupClient2" # Client = backuphost-12-fd # JobDefs = "DefaultJob" #} # Backup the catalog database (after the nightly save) Job { Name = "BackupCatalog" JobDefs = "DefaultFileJob" Level = Full FileSet="Catalog" Schedule = "WeeklyCycleAfterBackup" # This creates an ASCII copy of the catalog # Arguments to make_catalog_backup.pl are: # make_catalog_backup.pl <catalog-name> RunBeforeJob = "/etc/bacula/scripts/make_catalog_backup.pl MyCatalog" # This deletes the copy of the catalog RunAfterJob = "/etc/bacula/scripts/delete_catalog_backup" Write Bootstrap = "/var/lib/bacula/%n.bsr" Priority = 11 # run after main backup } # # Standard Restore template, to be changed by Console program # Only one such job is needed for all Jobs/Clients/Storage ... # Job { Name = "RestoreFiles" Type = Restore Client=backuphost-1-fd FileSet="Full Set" Storage = File Pool = Default Messages = Standard Where = /srv/bacula/restore } FileSet { Name = "SpecialVolume" Include { Options { signature = MD5 } File = /mnt/SpecialVolume } Exclude { File = /var/lib/bacula File = /nonexistant/path/to/file/archive/dir File = /proc File = /tmp File = /.journal File = /.fsck } } # List of files to be backed up FileSet { Name = "Full Set" Include { Options { signature = MD5 } File = /usr/sbin } Exclude { File = /var/lib/bacula File = /nonexistant/path/to/file/archive/dir File = /proc File = /tmp File = /.journal File = /.fsck } } Schedule { Name = "WeeklyCycle" Run = Full 1st sun at 23:05 Run = Differential 2nd-5th sun at 23:05 Run = Incremental mon-sat at 23:05 } # This schedule does the catalog. It starts after the WeeklyCycle Schedule { Name = "WeeklyCycleAfterBackup" Run = Full sun-sat at 23:10 } # This is the backup of the catalog FileSet { Name = "Catalog" Include { Options { signature = MD5 } File = "/var/lib/bacula/bacula.sql" } } # Client (File Services) to backup Client { Name = backuphost-1-fd Address = localhost FDPort = 9102 Catalog = MyCatalog Password = "surelyyourejoking" # password for FileDaemon File Retention = 30 days # 30 days Job Retention = 6 months # six months AutoPrune = yes # Prune expired Jobs/Files } # # Second Client (File Services) to backup # You should change Name, Address, and Password before using # #Client { # Name = backuphost-12-fd # Address = localhost2 # FDPort = 9102 # Catalog = MyCatalog # Password = "i'mnotjokinganddontcallmeshirley" # password for FileDaemon 2 # File Retention = 30 days # 30 days # Job Retention = 6 months # six months # AutoPrune = yes # Prune expired Jobs/Files #} # Definition of file storage device Storage { Name = File # Do not use "localhost" here Address = localhost # N.B. Use a fully qualified name here SDPort = 9103 Password = "lalalalala" Device = FileStorage Media Type = File } Storage { Name = "SpectraLogic" Address = localhost SDPort = 9103 Password = "linkedinmakethebestpasswords" Device = Drive-1 Device = Drive-2 Media Type = LTO5 Autochanger = yes } # Generic catalog service Catalog { Name = MyCatalog # Uncomment the following line if you want the dbi driver # dbdriver = "dbi:sqlite3"; dbaddress = 127.0.0.1; dbport = dbname = "bacula"; DB Address = ""; dbuser = "bacula"; dbpassword = "bbmaster63" } # Reasonable message delivery -- send most everything to email address # and to the console Messages { Name = Standard mailcommand = "/usr/lib/bacula/bsmtp -h localhost -f \"\(Bacula\) \<%r\>\" -s \"Bacula: %t %e of %c %l\" %r" operatorcommand = "/usr/lib/bacula/bsmtp -h localhost -f \"\(Bacula\) \<%r\>\" -s \"Bacula: Intervention needed for %j\" %r" mail = root@localhost = all, !skipped operator = root@localhost = mount console = all, !skipped, !saved # # WARNING! the following will create a file that you must cycle from # time to time as it will grow indefinitely. However, it will # also keep all your messages if they scroll off the console. # append = "/var/lib/bacula/log" = all, !skipped catalog = all } # # Message delivery for daemon messages (no job). Messages { Name = Daemon mailcommand = "/usr/lib/bacula/bsmtp -h localhost -f \"\(Bacula\) \<%r\>\" -s \"Bacula daemon message\" %r" mail = root@localhost = all, !skipped console = all, !skipped, !saved append = "/var/lib/bacula/log" = all, !skipped } # Default pool definition Pool { Name = Default Pool Type = Backup Recycle = yes # Bacula can automatically recycle Volumes AutoPrune = yes # Prune expired volumes Volume Retention = 365 days # one year } # File Pool definition Pool { Name = File Pool Type = Backup Recycle = yes # Bacula can automatically recycle Volumes AutoPrune = yes # Prune expired volumes Volume Retention = 365 days # one year Maximum Volume Bytes = 50G # Limit Volume size to something reasonable Maximum Volumes = 100 # Limit number of Volumes in Pool } Pool { Name = AllTapes Pool Type = Backup Recycle = yes AutoPrune = yes # Prune expired volumes Volume Retention = 31 days # one Moth } # Scratch pool definition Pool { Name = Scratch Pool Type = Backup } # # Restricted console used by tray-monitor to get the status of the director # Console { Name = backuphost-1-mon Password = "LastFMalsostorePasswordsLikeThis" CommandACL = status, .status } bacula-sd.conf # # Default Bacula Storage Daemon Configuration file # Storage { # definition of myself Name = backuphost-1-sd SDPort = 9103 # Director's port WorkingDirectory = "/var/lib/bacula" Pid Directory = "/var/run/bacula" Maximum Concurrent Jobs = 20 SDAddress = 0.0.0.0 # SDAddress = 127.0.0.1 } # # List Directors who are permitted to contact Storage daemon # Director { Name = backuphost-1-dir Password = "passwordslinplaintext" } # # Restricted Director, used by tray-monitor to get the # status of the storage daemon # Director { Name = backuphost-1-mon Password = "totalinsecurityabound" Monitor = yes } Device { Name = FileStorage Media Type = File Archive Device = /srv/bacula/archive LabelMedia = yes; # lets Bacula label unlabeled media Random Access = Yes; AutomaticMount = yes; # when device opened, read it RemovableMedia = no; AlwaysOpen = no; } Autochanger { Name = SpectraLogic Device = Drive-1 Device = Drive-2 Changer Command = "/etc/bacula/scripts/mtx-changer %c %o %S %a %d" Changer Device = /dev/sg4 } Device { Name = Drive-1 Drive Index = 0 Archive Device = /dev/nst0 Changer Device = /dev/sg4 Media Type = LTO5 AutoChanger = yes RemovableMedia = yes; AutomaticMount = yes; AlwaysOpen = yes; RandomAccess = no; LabelMedia = yes } Device { Name = Drive-2 Drive Index = 1 Archive Device = /dev/nst1 Changer Device = /dev/sg4 Media Type = LTO5 AutoChanger = yes RemovableMedia = yes; AutomaticMount = yes; AlwaysOpen = yes; RandomAccess = no; LabelMedia = yes } # # Send all messages to the Director, # mount messages also are sent to the email address # Messages { Name = Standard director = backuphost-1-dir = all } bacula-fd.conf # # Default Bacula File Daemon Configuration file # # # List Directors who are permitted to contact this File daemon # Director { Name = backuphost-1-dir Password = "hahahahahaha" } # # Restricted Director, used by tray-monitor to get the # status of the file daemon # Director { Name = backuphost-1-mon Password = "hohohohohho" Monitor = yes } # # "Global" File daemon configuration specifications # FileDaemon { # this is me Name = backuphost-1-fd FDport = 9102 # where we listen for the director WorkingDirectory = /var/lib/bacula Pid Directory = /var/run/bacula Maximum Concurrent Jobs = 20 #FDAddress = 127.0.0.1 FDAddress = 0.0.0.0 } # Send all messages except skipped files back to Director Messages { Name = Standard director = backuphost-1-dir = all, !skipped, !restored }

    Read the article

  • Authoritative sources about Database vs. Flatfile decision

    - by FastAl
    <tldr>looking for a reference to a book or other undeniably authoritative source that gives reasons when you should choose a database vs. when you should choose other storage methods. I have provided an un-authoritative list of reasons about 2/3 of the way down this post.</tldr> I have a situation at my company where a database is being used where it would be better to use another solution (in this case, an auto-generated piece of source code that contains a static lookup table, searched by binary sort). Normally, a database would be an OK solution even though the problem does not require a database, e.g, none of the elements of ACID are needed, as it is read-only data, updated about every 3-5 years (also requiring other sourcecode changes), and fits in memory, and can be keyed into via binary search (a tad faster than db, but speed is not an issue). The problem is that this code runs on our enterprise server, but is shared with several PC platforms (some disconnected, some use a central DB, etc.), and parts of it are managed by multiple programming units, parts by the DBAs, parts even by mathematicians in another department, etc. These hit their own platform’s version of their databases (containing their own copy of the static data). What happens is that every implementation, every little change, something different goes wrong. There are many other issues as well. I can’t even use a flatfile, because one mode of running on our enterprise server does not have permission to read files (only databases, and of course, its own literal storage, e.g., in-source table). Of course, other parts of the system use databases in proper, less obscure manners; there is no problem with those parts. So why don’t we just change it? I don’t have administrative ability to force a change. But I’m affected because sometimes I have to help fix the problems, but mostly because it causes outages and tons of extra IT time by other programmers and d*mmit that makes me mad! The reason neither management, nor the designers of the system, can see the problem is that they propose a solution that won’t work: increase communication; implement more safeguards and standards; etc. But every time, in a different part of the already-pared-down but still multi-step processes, a few different diligent, hard-working, top performing IT personnel make a unique subtle error that causes it to fail, sometimes after the last round of testing! And in general these are not single-person failures, but understandable miscommunications. And communication at our company is actually better than most. People just don't think that's the case because they haven't dug into the matter. However, I have it on very good word from somebody with extensive formal study of sociology and psychology that the relatively small amount of less-than-proper database usage in this gigantic cross-platform multi-source, multi-language project is bureaucratically un-maintainable. Impossible. No chance. At least with Human Beings in the loop, and it can’t be automated. In addition, the management and developers who could change this, though intelligent and capable, don’t understand the rigidity of this ‘how humans are’ issue, and are not convincible on the matter. The reason putting the static data in sourcecode will solve the problem is, although the solution is less sexy than a database, it would function with no technical drawbacks; and since the sharing of sourcecode already works very well, you basically erase any database-related effort from this section of the project, along with all the drawbacks of it that are causing problems. OK, that’s the background, for the curious. I won’t be able to convince management that this is an unfixable sociological problem, and that the real solution is coding around these limits of human nature, just as you would code around a bug in a 3rd party component that you can’t change. So what I have to do is exploit the unsuitableness of the database solution, and not do it using logic, but rather authority. I am aware of many reasons, and posts on this site giving reasons for one over the other; I’m not looking for lists of reasons like these (although you can add a comment if I've miss a doozy): WHY USE A DATABASE? instead of flatfile/other DB vs. file: if you need... Random Read / Transparent search optimization Advanced / varied / customizable Searching and sorting capabilities Transaction/rollback Locks, semaphores Concurrency control / Shared users Security 1-many/m-m is easier Easy modification Scalability Load Balancing Random updates / inserts / deletes Advanced query Administrative control of design, etc. SQL / learning curve Debugging / Logging Centralized / Live Backup capabilities Cached queries / dvlp & cache execution plans Interleaved update/read Referential integrity, avoid redundant/missing/corrupt/out-of-sync data Reporting (from on olap or oltp db) / turnkey generation tools [Disadvantages:] Important to get right the first time - professional design - but only b/c it's meant to last s/w & h/w cost Usu. over a network, speed issue (best vs. best design vs. local=even then a separate process req's marshalling/netwk layers/inter-p comm) indicies and query processing can stand in the way of simple processing (vs. flatfile) WHY USE FLATFILE: If you only need... Sequential Row processing only Limited usage append only (no reading, no master key/update) Only Update the record you're reading (fixed length recs only) Too big to fit into memory If Local disk / read-ahead network connection Portability / small system Email / cut & Paste / store as document by novice - simple format Low design learning curve but high cost later WHY USE IN-MEMORY/TABLE (tables, arrays, etc.): if you need... Processing a single db/ff record that was imported Known size of data Static data if hardcoding the table Narrow, unchanging use (e.g., one program or proc) -includes a class that will be shared, but encapsulates its data manipulation Extreme speed needed / high transaction frequency Random access - but search is dependent on implementation Following are some other posts about the topic: http://stackoverflow.com/questions/1499239/database-vs-flat-text-file-what-are-some-technical-reasons-for-choosing-one-over http://stackoverflow.com/questions/332825/are-flat-file-databases-any-good http://stackoverflow.com/questions/2356851/database-vs-flat-files http://stackoverflow.com/questions/514455/databases-vs-plain-text/514530 What I’d like to know is if anybody could recommend a hard, authoritative source containing these reasons. I’m looking for a paper book I can buy, or a reputable website with whitepapers about the issue (e.g., Microsoft, IBM), not counting the user-generated content on those sites. This will have a greater change to elicit a change that I’m looking for: less wasted programmer time, and more reliable programs. Thanks very much for your help. You win a prize for reading such a large post!

    Read the article

  • How do I prove I should put a table of values in source code instead of a database table?

    - by FastAl
    <tldr>looking for a reference to a book or other undeniably authoritative source that gives reasons when you should choose a database vs. when you should choose other storage methods. I have provided an un-authoritative list of reasons about 2/3 of the way down this post.</tldr> I have a situation at my company where a database is being used where it would be better to use another solution (in this case, an auto-generated piece of source code that contains a static lookup table, searched by binary sort). Normally, a database would be an OK solution even though the problem does not require a database, e.g, none of the elements of ACID are needed, as it is read-only data, updated about every 3-5 years (also requiring other sourcecode changes), and fits in memory, and can be keyed into via binary search (a tad faster than db, but speed is not an issue). The problem is that this code runs on our enterprise server, but is shared with several PC platforms (some disconnected, some use a central DB, etc.), and parts of it are managed by multiple programming units, parts by the DBAs, parts even by mathematicians in another department, etc. These hit their own platform’s version of their databases (containing their own copy of the static data). What happens is that every implementation, every little change, something different goes wrong. There are many other issues as well. I can’t even use a flatfile, because one mode of running on our enterprise server does not have permission to read files (only databases, and of course, its own literal storage, e.g., in-source table). Of course, other parts of the system use databases in proper, less obscure manners; there is no problem with those parts. So why don’t we just change it? I don’t have administrative ability to force a change. But I’m affected because sometimes I have to help fix the problems, but mostly because it causes outages and tons of extra IT time by other programmers and d*mmit that makes me mad! The reason neither management, nor the designers of the system, can see the problem is that they propose a solution that won’t work: increase communication; implement more safeguards and standards; etc. But every time, in a different part of the already-pared-down but still multi-step processes, a few different diligent, hard-working, top performing IT personnel make a unique subtle error that causes it to fail, sometimes after the last round of testing! And in general these are not single-person failures, but understandable miscommunications. And communication at our company is actually better than most. People just don't think that's the case because they haven't dug into the matter. However, I have it on very good word from somebody with extensive formal study of sociology and psychology that the relatively small amount of less-than-proper database usage in this gigantic cross-platform multi-source, multi-language project is bureaucratically un-maintainable. Impossible. No chance. At least with Human Beings in the loop, and it can’t be automated. In addition, the management and developers who could change this, though intelligent and capable, don’t understand the rigidity of this ‘how humans are’ issue, and are not convincible on the matter. The reason putting the static data in sourcecode will solve the problem is, although the solution is less sexy than a database, it would function with no technical drawbacks; and since the sharing of sourcecode already works very well, you basically erase any database-related effort from this section of the project, along with all the drawbacks of it that are causing problems. OK, that’s the background, for the curious. I won’t be able to convince management that this is an unfixable sociological problem, and that the real solution is coding around these limits of human nature, just as you would code around a bug in a 3rd party component that you can’t change. So what I have to do is exploit the unsuitableness of the database solution, and not do it using logic, but rather authority. I am aware of many reasons, and posts on this site giving reasons for one over the other; I’m not looking for lists of reasons like these (although you can add a comment if I've miss a doozy): WHY USE A DATABASE? instead of flatfile/other DB vs. file: if you need... Random Read / Transparent search optimization Advanced / varied / customizable Searching and sorting capabilities Transaction/rollback Locks, semaphores Concurrency control / Shared users Security 1-many/m-m is easier Easy modification Scalability Load Balancing Random updates / inserts / deletes Advanced query Administrative control of design, etc. SQL / learning curve Debugging / Logging Centralized / Live Backup capabilities Cached queries / dvlp & cache execution plans Interleaved update/read Referential integrity, avoid redundant/missing/corrupt/out-of-sync data Reporting (from on olap or oltp db) / turnkey generation tools [Disadvantages:] Important to get right the first time - professional design - but only b/c it's meant to last s/w & h/w cost Usu. over a network, speed issue (best vs. best design vs. local=even then a separate process req's marshalling/netwk layers/inter-p comm) indicies and query processing can stand in the way of simple processing (vs. flatfile) WHY USE FLATFILE: If you only need... Sequential Row processing only Limited usage append only (no reading, no master key/update) Only Update the record you're reading (fixed length recs only) Too big to fit into memory If Local disk / read-ahead network connection Portability / small system Email / cut & Paste / store as document by novice - simple format Low design learning curve but high cost later WHY USE IN-MEMORY/TABLE (tables, arrays, etc.): if you need... Processing a single db/ff record that was imported Known size of data Static data if hardcoding the table Narrow, unchanging use (e.g., one program or proc) -includes a class that will be shared, but encapsulates its data manipulation Extreme speed needed / high transaction frequency Random access - but search is dependent on implementation Following are some other posts about the topic: http://stackoverflow.com/questions/1499239/database-vs-flat-text-file-what-are-some-technical-reasons-for-choosing-one-over http://stackoverflow.com/questions/332825/are-flat-file-databases-any-good http://stackoverflow.com/questions/2356851/database-vs-flat-files http://stackoverflow.com/questions/514455/databases-vs-plain-text/514530 What I’d like to know is if anybody could recommend a hard, authoritative source containing these reasons. I’m looking for a paper book I can buy, or a reputable website with whitepapers about the issue (e.g., Microsoft, IBM), not counting the user-generated content on those sites. This will have a greater change to elicit a change that I’m looking for: less wasted programmer time, and more reliable programs. Thanks very much for your help. You win a prize for reading such a large post!

    Read the article

  • OpenCL: Strange buffer or image bahaviour with NVidia but not Amd

    - by Alex R.
    I have a big problem (on Linux): I create a buffer with defined data, then an OpenCL kernel takes this data and puts it into an image2d_t. When working on an AMD C50 (Fusion CPU/GPU) the program works as desired, but on my GeForce 9500 GT the given kernel computes the correct result very rarely. Sometimes the result is correct, but very often it is incorrect. Sometimes it depends on very strange changes like removing unused variable declarations or adding a newline. I realized that disabling the optimization will increase the probability to fail. I have the most actual display driver in both systems. Here is my reduced code: #include <CL/cl.h> #include <string> #include <iostream> #include <sstream> #include <cmath> void checkOpenCLErr(cl_int err, std::string name){ const char* errorString[] = { "CL_SUCCESS", "CL_DEVICE_NOT_FOUND", "CL_DEVICE_NOT_AVAILABLE", "CL_COMPILER_NOT_AVAILABLE", "CL_MEM_OBJECT_ALLOCATION_FAILURE", "CL_OUT_OF_RESOURCES", "CL_OUT_OF_HOST_MEMORY", "CL_PROFILING_INFO_NOT_AVAILABLE", "CL_MEM_COPY_OVERLAP", "CL_IMAGE_FORMAT_MISMATCH", "CL_IMAGE_FORMAT_NOT_SUPPORTED", "CL_BUILD_PROGRAM_FAILURE", "CL_MAP_FAILURE", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "CL_INVALID_VALUE", "CL_INVALID_DEVICE_TYPE", "CL_INVALID_PLATFORM", "CL_INVALID_DEVICE", "CL_INVALID_CONTEXT", "CL_INVALID_QUEUE_PROPERTIES", "CL_INVALID_COMMAND_QUEUE", "CL_INVALID_HOST_PTR", "CL_INVALID_MEM_OBJECT", "CL_INVALID_IMAGE_FORMAT_DESCRIPTOR", "CL_INVALID_IMAGE_SIZE", "CL_INVALID_SAMPLER", "CL_INVALID_BINARY", "CL_INVALID_BUILD_OPTIONS", "CL_INVALID_PROGRAM", "CL_INVALID_PROGRAM_EXECUTABLE", "CL_INVALID_KERNEL_NAME", "CL_INVALID_KERNEL_DEFINITION", "CL_INVALID_KERNEL", "CL_INVALID_ARG_INDEX", "CL_INVALID_ARG_VALUE", "CL_INVALID_ARG_SIZE", "CL_INVALID_KERNEL_ARGS", "CL_INVALID_WORK_DIMENSION", "CL_INVALID_WORK_GROUP_SIZE", "CL_INVALID_WORK_ITEM_SIZE", "CL_INVALID_GLOBAL_OFFSET", "CL_INVALID_EVENT_WAIT_LIST", "CL_INVALID_EVENT", "CL_INVALID_OPERATION", "CL_INVALID_GL_OBJECT", "CL_INVALID_BUFFER_SIZE", "CL_INVALID_MIP_LEVEL", "CL_INVALID_GLOBAL_WORK_SIZE", }; if (err != CL_SUCCESS) { std::stringstream str; str << errorString[-err] << " (" << err << ")"; throw std::string(name)+(str.str()); } } int main(){ try{ cl_context m_context; cl_platform_id* m_platforms; unsigned int m_numPlatforms; cl_command_queue m_queue; cl_device_id m_device; cl_int error = 0; // Used to handle error codes clGetPlatformIDs(0,NULL,&m_numPlatforms); m_platforms = new cl_platform_id[m_numPlatforms]; error = clGetPlatformIDs(m_numPlatforms,m_platforms,&m_numPlatforms); checkOpenCLErr(error, "getPlatformIDs"); // Device error = clGetDeviceIDs(m_platforms[0], CL_DEVICE_TYPE_GPU, 1, &m_device, NULL); checkOpenCLErr(error, "getDeviceIDs"); // Context cl_context_properties properties[] = { CL_CONTEXT_PLATFORM, (cl_context_properties)(m_platforms[0]), 0}; m_context = clCreateContextFromType(properties, CL_DEVICE_TYPE_GPU, NULL, NULL, NULL); // m_private->m_context = clCreateContext(properties, 1, &m_private->m_device, NULL, NULL, &error); checkOpenCLErr(error, "Create context"); // Command-queue m_queue = clCreateCommandQueue(m_context, m_device, 0, &error); checkOpenCLErr(error, "Create command queue"); //Build program and kernel const char* source = "#pragma OPENCL EXTENSION cl_khr_byte_addressable_store : enable\n" "\n" "__kernel void bufToImage(__global unsigned char* in, __write_only image2d_t out, const unsigned int offset_x, const unsigned int image_width , const unsigned int maxval ){\n" "\tint i = get_global_id(0);\n" "\tint j = get_global_id(1);\n" "\tint width = get_global_size(0);\n" "\tint height = get_global_size(1);\n" "\n" "\tint pos = j*image_width*3+(offset_x+i)*3;\n" "\tif( maxval < 256 ){\n" "\t\tfloat4 c = (float4)(in[pos],in[pos+1],in[pos+2],1.0f);\n" "\t\tc.x /= maxval;\n" "\t\tc.y /= maxval;\n" "\t\tc.z /= maxval;\n" "\t\twrite_imagef(out, (int2)(i,j), c);\n" "\t}else{\n" "\t\tfloat4 c = (float4)(255.0f*in[2*pos]+in[2*pos+1],255.0f*in[2*pos+2]+in[2*pos+3],255.0f*in[2*pos+4]+in[2*pos+5],1.0f);\n" "\t\tc.x /= maxval;\n" "\t\tc.y /= maxval;\n" "\t\tc.z /= maxval;\n" "\t\twrite_imagef(out, (int2)(i,j), c);\n" "\t}\n" "}\n" "\n" "__constant sampler_t imageSampler = CLK_NORMALIZED_COORDS_FALSE | CLK_ADDRESS_CLAMP_TO_EDGE | CLK_FILTER_NEAREST;\n" "\n" "__kernel void imageToBuf(__read_only image2d_t in, __global unsigned char* out, const unsigned int offset_x, const unsigned int image_width ){\n" "\tint i = get_global_id(0);\n" "\tint j = get_global_id(1);\n" "\tint pos = j*image_width*3+(offset_x+i)*3;\n" "\tfloat4 c = read_imagef(in, imageSampler, (int2)(i,j));\n" "\tif( c.x <= 1.0f && c.y <= 1.0f && c.z <= 1.0f ){\n" "\t\tout[pos] = c.x*255.0f;\n" "\t\tout[pos+1] = c.y*255.0f;\n" "\t\tout[pos+2] = c.z*255.0f;\n" "\t}else{\n" "\t\tout[pos] = 200.0f;\n" "\t\tout[pos+1] = 0.0f;\n" "\t\tout[pos+2] = 255.0f;\n" "\t}\n" "}\n"; cl_int err; cl_program prog = clCreateProgramWithSource(m_context,1,&source,NULL,&err); if( -err != CL_SUCCESS ) throw std::string("clCreateProgramWithSources"); err = clBuildProgram(prog,0,NULL,"-cl-opt-disable",NULL,NULL); if( -err != CL_SUCCESS ) throw std::string("clBuildProgram(fromSources)"); cl_kernel kernel = clCreateKernel(prog,"bufToImage",&err); checkOpenCLErr(err,"CreateKernel"); cl_uint imageWidth = 8; cl_uint imageHeight = 9; //Initialize datas cl_uint maxVal = 255; cl_uint offsetX = 0; int size = imageWidth*imageHeight*3; int resSize = imageWidth*imageHeight*4; cl_uchar* data = new cl_uchar[size]; cl_float* expectedData = new cl_float[resSize]; for( int i = 0,j=0; i < size; i++,j++ ){ data[i] = (cl_uchar)i; expectedData[j] = (cl_float)i/255.0f; if ( i%3 == 2 ){ j++; expectedData[j] = 1.0f; } } cl_mem inBuffer = clCreateBuffer(m_context,CL_MEM_READ_ONLY|CL_MEM_COPY_HOST_PTR,size*sizeof(cl_uchar),data,&err); checkOpenCLErr(err, "clCreateBuffer()"); clFinish(m_queue); cl_image_format imgFormat; imgFormat.image_channel_order = CL_RGBA; imgFormat.image_channel_data_type = CL_FLOAT; cl_mem outImg = clCreateImage2D( m_context, CL_MEM_READ_WRITE, &imgFormat, imageWidth, imageHeight, 0, NULL, &err ); checkOpenCLErr(err,"get2DImage()"); clFinish(m_queue); size_t kernelRegion[]={imageWidth,imageHeight}; size_t kernelWorkgroup[]={1,1}; //Fill kernel with data clSetKernelArg(kernel,0,sizeof(cl_mem),&inBuffer); clSetKernelArg(kernel,1,sizeof(cl_mem),&outImg); clSetKernelArg(kernel,2,sizeof(cl_uint),&offsetX); clSetKernelArg(kernel,3,sizeof(cl_uint),&imageWidth); clSetKernelArg(kernel,4,sizeof(cl_uint),&maxVal); //Run kernel err = clEnqueueNDRangeKernel(m_queue,kernel,2,NULL,kernelRegion,kernelWorkgroup,0,NULL,NULL); checkOpenCLErr(err,"RunKernel"); clFinish(m_queue); //Check resulting data for validty cl_float* computedData = new cl_float[resSize];; size_t region[]={imageWidth,imageHeight,1}; const size_t offset[] = {0,0,0}; err = clEnqueueReadImage(m_queue,outImg,CL_TRUE,offset,region,0,0,computedData,0,NULL,NULL); checkOpenCLErr(err, "readDataFromImage()"); clFinish(m_queue); for( int i = 0; i < resSize; i++ ){ if( fabs(expectedData[i]-computedData[i])>0.1 ){ std::cout << "Expected: \n"; for( int j = 0; j < resSize; j++ ){ std::cout << expectedData[j] << " "; } std::cout << "\nComputed: \n"; std::cout << "\n"; for( int j = 0; j < resSize; j++ ){ std::cout << computedData[j] << " "; } std::cout << "\n"; throw std::string("Error, computed and expected data are not the same!\n"); } } }catch(std::string& e){ std::cout << "\nCaught an exception: " << e << "\n"; return 1; } std::cout << "Works fine\n"; return 0; } I also uploaded the source code for you to make it easier to test it: http://www.file-upload.net/download-3513797/strangeOpenCLError.cpp.html Please can you tell me if I've done wrong anything? Is there any mistake in the code or is this a bug in my driver? Best reagards, Alex

    Read the article

  • Using R to Analyze G1GC Log Files

    - by user12620111
    Using R to Analyze G1GC Log Files body, td { font-family: sans-serif; background-color: white; font-size: 12px; margin: 8px; } tt, code, pre { font-family: 'DejaVu Sans Mono', 'Droid Sans Mono', 'Lucida Console', Consolas, Monaco, monospace; } h1 { font-size:2.2em; } h2 { font-size:1.8em; } h3 { font-size:1.4em; } h4 { font-size:1.0em; } h5 { font-size:0.9em; } h6 { font-size:0.8em; } a:visited { color: rgb(50%, 0%, 50%); } pre { margin-top: 0; max-width: 95%; border: 1px solid #ccc; white-space: pre-wrap; } pre code { display: block; padding: 0.5em; } code.r, code.cpp { background-color: #F8F8F8; } table, td, th { border: none; } blockquote { color:#666666; margin:0; padding-left: 1em; border-left: 0.5em #EEE solid; } hr { height: 0px; border-bottom: none; border-top-width: thin; border-top-style: dotted; border-top-color: #999999; } @media print { * { background: transparent !important; color: black !important; filter:none !important; -ms-filter: none !important; } body { font-size:12pt; max-width:100%; } a, a:visited { text-decoration: underline; } hr { visibility: hidden; page-break-before: always; } pre, blockquote { padding-right: 1em; page-break-inside: avoid; } tr, img { page-break-inside: avoid; } img { max-width: 100% !important; } @page :left { margin: 15mm 20mm 15mm 10mm; } @page :right { margin: 15mm 10mm 15mm 20mm; } p, h2, h3 { orphans: 3; widows: 3; } h2, h3 { page-break-after: avoid; } } pre .operator, pre .paren { color: rgb(104, 118, 135) } pre .literal { color: rgb(88, 72, 246) } pre .number { color: rgb(0, 0, 205); } pre .comment { color: rgb(76, 136, 107); } pre .keyword { color: rgb(0, 0, 255); } pre .identifier { color: rgb(0, 0, 0); } pre .string { color: rgb(3, 106, 7); } var hljs=new function(){function m(p){return p.replace(/&/gm,"&").replace(/"}while(y.length||w.length){var v=u().splice(0,1)[0];z+=m(x.substr(q,v.offset-q));q=v.offset;if(v.event=="start"){z+=t(v.node);s.push(v.node)}else{if(v.event=="stop"){var p,r=s.length;do{r--;p=s[r];z+=("")}while(p!=v.node);s.splice(r,1);while(r'+M[0]+""}else{r+=M[0]}O=P.lR.lastIndex;M=P.lR.exec(L)}return r+L.substr(O,L.length-O)}function J(L,M){if(M.sL&&e[M.sL]){var r=d(M.sL,L);x+=r.keyword_count;return r.value}else{return F(L,M)}}function I(M,r){var L=M.cN?'':"";if(M.rB){y+=L;M.buffer=""}else{if(M.eB){y+=m(r)+L;M.buffer=""}else{y+=L;M.buffer=r}}D.push(M);A+=M.r}function G(N,M,Q){var R=D[D.length-1];if(Q){y+=J(R.buffer+N,R);return false}var P=q(M,R);if(P){y+=J(R.buffer+N,R);I(P,M);return P.rB}var L=v(D.length-1,M);if(L){var O=R.cN?"":"";if(R.rE){y+=J(R.buffer+N,R)+O}else{if(R.eE){y+=J(R.buffer+N,R)+O+m(M)}else{y+=J(R.buffer+N+M,R)+O}}while(L1){O=D[D.length-2].cN?"":"";y+=O;L--;D.length--}var r=D[D.length-1];D.length--;D[D.length-1].buffer="";if(r.starts){I(r.starts,"")}return R.rE}if(w(M,R)){throw"Illegal"}}var E=e[B];var D=[E.dM];var A=0;var x=0;var y="";try{var s,u=0;E.dM.buffer="";do{s=p(C,u);var t=G(s[0],s[1],s[2]);u+=s[0].length;if(!t){u+=s[1].length}}while(!s[2]);if(D.length1){throw"Illegal"}return{r:A,keyword_count:x,value:y}}catch(H){if(H=="Illegal"){return{r:0,keyword_count:0,value:m(C)}}else{throw H}}}function g(t){var p={keyword_count:0,r:0,value:m(t)};var r=p;for(var q in e){if(!e.hasOwnProperty(q)){continue}var s=d(q,t);s.language=q;if(s.keyword_count+s.rr.keyword_count+r.r){r=s}if(s.keyword_count+s.rp.keyword_count+p.r){r=p;p=s}}if(r.language){p.second_best=r}return p}function i(r,q,p){if(q){r=r.replace(/^((]+|\t)+)/gm,function(t,w,v,u){return w.replace(/\t/g,q)})}if(p){r=r.replace(/\n/g,"")}return r}function n(t,w,r){var x=h(t,r);var v=a(t);var y,s;if(v){y=d(v,x)}else{return}var q=c(t);if(q.length){s=document.createElement("pre");s.innerHTML=y.value;y.value=k(q,c(s),x)}y.value=i(y.value,w,r);var u=t.className;if(!u.match("(\\s|^)(language-)?"+v+"(\\s|$)")){u=u?(u+" "+v):v}if(/MSIE [678]/.test(navigator.userAgent)&&t.tagName=="CODE"&&t.parentNode.tagName=="PRE"){s=t.parentNode;var p=document.createElement("div");p.innerHTML=""+y.value+"";t=p.firstChild.firstChild;p.firstChild.cN=s.cN;s.parentNode.replaceChild(p.firstChild,s)}else{t.innerHTML=y.value}t.className=u;t.result={language:v,kw:y.keyword_count,re:y.r};if(y.second_best){t.second_best={language:y.second_best.language,kw:y.second_best.keyword_count,re:y.second_best.r}}}function o(){if(o.called){return}o.called=true;var r=document.getElementsByTagName("pre");for(var p=0;p|=||=||=|\\?|\\[|\\{|\\(|\\^|\\^=|\\||\\|=|\\|\\||~";this.ER="(?![\\s\\S])";this.BE={b:"\\\\.",r:0};this.ASM={cN:"string",b:"'",e:"'",i:"\\n",c:[this.BE],r:0};this.QSM={cN:"string",b:'"',e:'"',i:"\\n",c:[this.BE],r:0};this.CLCM={cN:"comment",b:"//",e:"$"};this.CBLCLM={cN:"comment",b:"/\\*",e:"\\*/"};this.HCM={cN:"comment",b:"#",e:"$"};this.NM={cN:"number",b:this.NR,r:0};this.CNM={cN:"number",b:this.CNR,r:0};this.BNM={cN:"number",b:this.BNR,r:0};this.inherit=function(r,s){var p={};for(var q in r){p[q]=r[q]}if(s){for(var q in s){p[q]=s[q]}}return p}}();hljs.LANGUAGES.cpp=function(){var a={keyword:{"false":1,"int":1,"float":1,"while":1,"private":1,"char":1,"catch":1,"export":1,virtual:1,operator:2,sizeof:2,dynamic_cast:2,typedef:2,const_cast:2,"const":1,struct:1,"for":1,static_cast:2,union:1,namespace:1,unsigned:1,"long":1,"throw":1,"volatile":2,"static":1,"protected":1,bool:1,template:1,mutable:1,"if":1,"public":1,friend:2,"do":1,"return":1,"goto":1,auto:1,"void":2,"enum":1,"else":1,"break":1,"new":1,extern:1,using:1,"true":1,"class":1,asm:1,"case":1,typeid:1,"short":1,reinterpret_cast:2,"default":1,"double":1,register:1,explicit:1,signed:1,typename:1,"try":1,"this":1,"switch":1,"continue":1,wchar_t:1,inline:1,"delete":1,alignof:1,char16_t:1,char32_t:1,constexpr:1,decltype:1,noexcept:1,nullptr:1,static_assert:1,thread_local:1,restrict:1,_Bool:1,complex:1},built_in:{std:1,string:1,cin:1,cout:1,cerr:1,clog:1,stringstream:1,istringstream:1,ostringstream:1,auto_ptr:1,deque:1,list:1,queue:1,stack:1,vector:1,map:1,set:1,bitset:1,multiset:1,multimap:1,unordered_set:1,unordered_map:1,unordered_multiset:1,unordered_multimap:1,array:1,shared_ptr:1}};return{dM:{k:a,i:"",k:a,r:10,c:["self"]}]}}}();hljs.LANGUAGES.r={dM:{c:[hljs.HCM,{cN:"number",b:"\\b0[xX][0-9a-fA-F]+[Li]?\\b",e:hljs.IMMEDIATE_RE,r:0},{cN:"number",b:"\\b\\d+(?:[eE][+\\-]?\\d*)?L\\b",e:hljs.IMMEDIATE_RE,r:0},{cN:"number",b:"\\b\\d+\\.(?!\\d)(?:i\\b)?",e:hljs.IMMEDIATE_RE,r:1},{cN:"number",b:"\\b\\d+(?:\\.\\d*)?(?:[eE][+\\-]?\\d*)?i?\\b",e:hljs.IMMEDIATE_RE,r:0},{cN:"number",b:"\\.\\d+(?:[eE][+\\-]?\\d*)?i?\\b",e:hljs.IMMEDIATE_RE,r:1},{cN:"keyword",b:"(?:tryCatch|library|setGeneric|setGroupGeneric)\\b",e:hljs.IMMEDIATE_RE,r:10},{cN:"keyword",b:"\\.\\.\\.",e:hljs.IMMEDIATE_RE,r:10},{cN:"keyword",b:"\\.\\.\\d+(?![\\w.])",e:hljs.IMMEDIATE_RE,r:10},{cN:"keyword",b:"\\b(?:function)",e:hljs.IMMEDIATE_RE,r:2},{cN:"keyword",b:"(?:if|in|break|next|repeat|else|for|return|switch|while|try|stop|warning|require|attach|detach|source|setMethod|setClass)\\b",e:hljs.IMMEDIATE_RE,r:1},{cN:"literal",b:"(?:NA|NA_integer_|NA_real_|NA_character_|NA_complex_)\\b",e:hljs.IMMEDIATE_RE,r:10},{cN:"literal",b:"(?:NULL|TRUE|FALSE|T|F|Inf|NaN)\\b",e:hljs.IMMEDIATE_RE,r:1},{cN:"identifier",b:"[a-zA-Z.][a-zA-Z0-9._]*\\b",e:hljs.IMMEDIATE_RE,r:0},{cN:"operator",b:"|=||   Using R to Analyze G1GC Log Files   Using R to Analyze G1GC Log Files Introduction Working in Oracle Platform Integration gives an engineer opportunities to work on a wide array of technologies. My team’s goal is to make Oracle applications run best on the Solaris/SPARC platform. When looking for bottlenecks in a modern applications, one needs to be aware of not only how the CPUs and operating system are executing, but also network, storage, and in some cases, the Java Virtual Machine. I was recently presented with about 1.5 GB of Java Garbage First Garbage Collector log file data. If you’re not familiar with the subject, you might want to review Garbage First Garbage Collector Tuning by Monica Beckwith. The customer had been running Java HotSpot 1.6.0_31 to host a web application server. I was told that the Solaris/SPARC server was running a Java process launched using a commmand line that included the following flags: -d64 -Xms9g -Xmx9g -XX:+UseG1GC -XX:MaxGCPauseMillis=200 -XX:InitiatingHeapOccupancyPercent=80 -XX:PermSize=256m -XX:MaxPermSize=256m -XX:+PrintGC -XX:+PrintGCTimeStamps -XX:+PrintHeapAtGC -XX:+PrintGCDateStamps -XX:+PrintFlagsFinal -XX:+DisableExplicitGC -XX:+UnlockExperimentalVMOptions -XX:ParallelGCThreads=8 Several sources on the internet indicate that if I were to print out the 1.5 GB of log files, it would require enough paper to fill the bed of a pick up truck. Of course, it would be fruitless to try to scan the log files by hand. Tools will be required to summarize the contents of the log files. Others have encountered large Java garbage collection log files. There are existing tools to analyze the log files: IBM’s GC toolkit The chewiebug GCViewer gchisto HPjmeter Instead of using one of the other tools listed, I decide to parse the log files with standard Unix tools, and analyze the data with R. Data Cleansing The log files arrived in two different formats. I guess that the difference is that one set of log files was generated using a more verbose option, maybe -XX:+PrintHeapAtGC, and the other set of log files was generated without that option. Format 1 In some of the log files, the log files with the less verbose format, a single trace, i.e. the report of a singe garbage collection event, looks like this: {Heap before GC invocations=12280 (full 61): garbage-first heap total 9437184K, used 7499918K [0xfffffffd00000000, 0xffffffff40000000, 0xffffffff40000000) region size 4096K, 1 young (4096K), 0 survivors (0K) compacting perm gen total 262144K, used 144077K [0xffffffff40000000, 0xffffffff50000000, 0xffffffff50000000) the space 262144K, 54% used [0xffffffff40000000, 0xffffffff48cb3758, 0xffffffff48cb3800, 0xffffffff50000000) No shared spaces configured. 2014-05-14T07:24:00.988-0700: 60586.353: [GC pause (young) 7324M->7320M(9216M), 0.1567265 secs] Heap after GC invocations=12281 (full 61): garbage-first heap total 9437184K, used 7496533K [0xfffffffd00000000, 0xffffffff40000000, 0xffffffff40000000) region size 4096K, 0 young (0K), 0 survivors (0K) compacting perm gen total 262144K, used 144077K [0xffffffff40000000, 0xffffffff50000000, 0xffffffff50000000) the space 262144K, 54% used [0xffffffff40000000, 0xffffffff48cb3758, 0xffffffff48cb3800, 0xffffffff50000000) No shared spaces configured. } A simple grep can be used to extract a summary: $ grep "\[ GC pause (young" g1gc.log 2014-05-13T13:24:35.091-0700: 3.109: [GC pause (young) 20M->5029K(9216M), 0.0146328 secs] 2014-05-13T13:24:35.440-0700: 3.459: [GC pause (young) 9125K->6077K(9216M), 0.0086723 secs] 2014-05-13T13:24:37.581-0700: 5.599: [GC pause (young) 25M->8470K(9216M), 0.0203820 secs] 2014-05-13T13:24:42.686-0700: 10.704: [GC pause (young) 44M->15M(9216M), 0.0288848 secs] 2014-05-13T13:24:48.941-0700: 16.958: [GC pause (young) 51M->20M(9216M), 0.0491244 secs] 2014-05-13T13:24:56.049-0700: 24.066: [GC pause (young) 92M->26M(9216M), 0.0525368 secs] 2014-05-13T13:25:34.368-0700: 62.383: [GC pause (young) 602M->68M(9216M), 0.1721173 secs] But that format wasn't easily read into R, so I needed to be a bit more tricky. I used the following Unix command to create a summary file that was easy for R to read. $ echo "SecondsSinceLaunch BeforeSize AfterSize TotalSize RealTime" $ grep "\[GC pause (young" g1gc.log | grep -v mark | sed -e 's/[A-SU-z\(\),]/ /g' -e 's/->/ /' -e 's/: / /g' | more SecondsSinceLaunch BeforeSize AfterSize TotalSize RealTime 2014-05-13T13:24:35.091-0700 3.109 20 5029 9216 0.0146328 2014-05-13T13:24:35.440-0700 3.459 9125 6077 9216 0.0086723 2014-05-13T13:24:37.581-0700 5.599 25 8470 9216 0.0203820 2014-05-13T13:24:42.686-0700 10.704 44 15 9216 0.0288848 2014-05-13T13:24:48.941-0700 16.958 51 20 9216 0.0491244 2014-05-13T13:24:56.049-0700 24.066 92 26 9216 0.0525368 2014-05-13T13:25:34.368-0700 62.383 602 68 9216 0.1721173 Format 2 In some of the log files, the log files with the more verbose format, a single trace, i.e. the report of a singe garbage collection event, was more complicated than Format 1. Here is a text file with an example of a single G1GC trace in the second format. As you can see, it is quite complicated. It is nice that there is so much information available, but the level of detail can be overwhelming. I wrote this awk script (download) to summarize each trace on a single line. #!/usr/bin/env awk -f BEGIN { printf("SecondsSinceLaunch IncrementalCount FullCount UserTime SysTime RealTime BeforeSize AfterSize TotalSize\n") } ###################### # Save count data from lines that are at the start of each G1GC trace. # Each trace starts out like this: # {Heap before GC invocations=14 (full 0): # garbage-first heap total 9437184K, used 325496K [0xfffffffd00000000, 0xffffffff40000000, 0xffffffff40000000) ###################### /{Heap.*full/{ gsub ( "\\)" , "" ); nf=split($0,a,"="); split(a[2],b," "); getline; if ( match($0, "first") ) { G1GC=1; IncrementalCount=b[1]; FullCount=substr( b[3], 1, length(b[3])-1 ); } else { G1GC=0; } } ###################### # Pull out time stamps that are in lines with this format: # 2014-05-12T14:02:06.025-0700: 94.312: [GC pause (young), 0.08870154 secs] ###################### /GC pause/ { DateTime=$1; SecondsSinceLaunch=substr($2, 1, length($2)-1); } ###################### # Heap sizes are in lines that look like this: # [ 4842M->4838M(9216M)] ###################### /\[ .*]$/ { gsub ( "\\[" , "" ); gsub ( "\ \]" , "" ); gsub ( "->" , " " ); gsub ( "\\( " , " " ); gsub ( "\ \)" , " " ); split($0,a," "); if ( split(a[1],b,"M") > 1 ) {BeforeSize=b[1]*1024;} if ( split(a[1],b,"K") > 1 ) {BeforeSize=b[1];} if ( split(a[2],b,"M") > 1 ) {AfterSize=b[1]*1024;} if ( split(a[2],b,"K") > 1 ) {AfterSize=b[1];} if ( split(a[3],b,"M") > 1 ) {TotalSize=b[1]*1024;} if ( split(a[3],b,"K") > 1 ) {TotalSize=b[1];} } ###################### # Emit an output line when you find input that looks like this: # [Times: user=1.41 sys=0.08, real=0.24 secs] ###################### /\[Times/ { if (G1GC==1) { gsub ( "," , "" ); split($2,a,"="); UserTime=a[2]; split($3,a,"="); SysTime=a[2]; split($4,a,"="); RealTime=a[2]; print DateTime,SecondsSinceLaunch,IncrementalCount,FullCount,UserTime,SysTime,RealTime,BeforeSize,AfterSize,TotalSize; G1GC=0; } } The resulting summary is about 25X smaller that the original file, but still difficult for a human to digest. SecondsSinceLaunch IncrementalCount FullCount UserTime SysTime RealTime BeforeSize AfterSize TotalSize ... 2014-05-12T18:36:34.669-0700: 3985.744 561 0 0.57 0.06 0.16 1724416 1720320 9437184 2014-05-12T18:36:34.839-0700: 3985.914 562 0 0.51 0.06 0.19 1724416 1720320 9437184 2014-05-12T18:36:35.069-0700: 3986.144 563 0 0.60 0.04 0.27 1724416 1721344 9437184 2014-05-12T18:36:35.354-0700: 3986.429 564 0 0.33 0.04 0.09 1725440 1722368 9437184 2014-05-12T18:36:35.545-0700: 3986.620 565 0 0.58 0.04 0.17 1726464 1722368 9437184 2014-05-12T18:36:35.726-0700: 3986.801 566 0 0.43 0.05 0.12 1726464 1722368 9437184 2014-05-12T18:36:35.856-0700: 3986.930 567 0 0.30 0.04 0.07 1726464 1723392 9437184 2014-05-12T18:36:35.947-0700: 3987.023 568 0 0.61 0.04 0.26 1727488 1723392 9437184 2014-05-12T18:36:36.228-0700: 3987.302 569 0 0.46 0.04 0.16 1731584 1724416 9437184 Reading the Data into R Once the GC log data had been cleansed, either by processing the first format with the shell script, or by processing the second format with the awk script, it was easy to read the data into R. g1gc.df = read.csv("summary.txt", row.names = NULL, stringsAsFactors=FALSE,sep="") str(g1gc.df) ## 'data.frame': 8307 obs. of 10 variables: ## $ row.names : chr "2014-05-12T14:00:32.868-0700:" "2014-05-12T14:00:33.179-0700:" "2014-05-12T14:00:33.677-0700:" "2014-05-12T14:00:35.538-0700:" ... ## $ SecondsSinceLaunch: num 1.16 1.47 1.97 3.83 6.1 ... ## $ IncrementalCount : int 0 1 2 3 4 5 6 7 8 9 ... ## $ FullCount : int 0 0 0 0 0 0 0 0 0 0 ... ## $ UserTime : num 0.11 0.05 0.04 0.21 0.08 0.26 0.31 0.33 0.34 0.56 ... ## $ SysTime : num 0.04 0.01 0.01 0.05 0.01 0.06 0.07 0.06 0.07 0.09 ... ## $ RealTime : num 0.02 0.02 0.01 0.04 0.02 0.04 0.05 0.04 0.04 0.06 ... ## $ BeforeSize : int 8192 5496 5768 22528 24576 43008 34816 53248 55296 93184 ... ## $ AfterSize : int 1400 1672 2557 4907 7072 14336 16384 18432 19456 21504 ... ## $ TotalSize : int 9437184 9437184 9437184 9437184 9437184 9437184 9437184 9437184 9437184 9437184 ... head(g1gc.df) ## row.names SecondsSinceLaunch IncrementalCount ## 1 2014-05-12T14:00:32.868-0700: 1.161 0 ## 2 2014-05-12T14:00:33.179-0700: 1.472 1 ## 3 2014-05-12T14:00:33.677-0700: 1.969 2 ## 4 2014-05-12T14:00:35.538-0700: 3.830 3 ## 5 2014-05-12T14:00:37.811-0700: 6.103 4 ## 6 2014-05-12T14:00:41.428-0700: 9.720 5 ## FullCount UserTime SysTime RealTime BeforeSize AfterSize TotalSize ## 1 0 0.11 0.04 0.02 8192 1400 9437184 ## 2 0 0.05 0.01 0.02 5496 1672 9437184 ## 3 0 0.04 0.01 0.01 5768 2557 9437184 ## 4 0 0.21 0.05 0.04 22528 4907 9437184 ## 5 0 0.08 0.01 0.02 24576 7072 9437184 ## 6 0 0.26 0.06 0.04 43008 14336 9437184 Basic Statistics Once the data has been read into R, simple statistics are very easy to generate. All of the numbers from high school statistics are available via simple commands. For example, generate a summary of every column: summary(g1gc.df) ## row.names SecondsSinceLaunch IncrementalCount FullCount ## Length:8307 Min. : 1 Min. : 0 Min. : 0.0 ## Class :character 1st Qu.: 9977 1st Qu.:2048 1st Qu.: 0.0 ## Mode :character Median :12855 Median :4136 Median : 12.0 ## Mean :12527 Mean :4156 Mean : 31.6 ## 3rd Qu.:15758 3rd Qu.:6262 3rd Qu.: 61.0 ## Max. :55484 Max. :8391 Max. :113.0 ## UserTime SysTime RealTime BeforeSize ## Min. :0.040 Min. :0.0000 Min. : 0.0 Min. : 5476 ## 1st Qu.:0.470 1st Qu.:0.0300 1st Qu.: 0.1 1st Qu.:5137920 ## Median :0.620 Median :0.0300 Median : 0.1 Median :6574080 ## Mean :0.751 Mean :0.0355 Mean : 0.3 Mean :5841855 ## 3rd Qu.:0.920 3rd Qu.:0.0400 3rd Qu.: 0.2 3rd Qu.:7084032 ## Max. :3.370 Max. :1.5600 Max. :488.1 Max. :8696832 ## AfterSize TotalSize ## Min. : 1380 Min. :9437184 ## 1st Qu.:5002752 1st Qu.:9437184 ## Median :6559744 Median :9437184 ## Mean :5785454 Mean :9437184 ## 3rd Qu.:7054336 3rd Qu.:9437184 ## Max. :8482816 Max. :9437184 Q: What is the total amount of User CPU time spent in garbage collection? sum(g1gc.df$UserTime) ## [1] 6236 As you can see, less than two hours of CPU time was spent in garbage collection. Is that too much? To find the percentage of time spent in garbage collection, divide the number above by total_elapsed_time*CPU_count. In this case, there are a lot of CPU’s and it turns out the the overall amount of CPU time spent in garbage collection isn’t a problem when viewed in isolation. When calculating rates, i.e. events per unit time, you need to ask yourself if the rate is homogenous across the time period in the log file. Does the log file include spikes of high activity that should be separately analyzed? Averaging in data from nights and weekends with data from business hours may alias problems. If you have a reason to suspect that the garbage collection rates include peaks and valleys that need independent analysis, see the “Time Series” section, below. Q: How much garbage is collected on each pass? The amount of heap space that is recovered per GC pass is surprisingly low: At least one collection didn’t recover any data. (“Min.=0”) 25% of the passes recovered 3MB or less. (“1st Qu.=3072”) Half of the GC passes recovered 4MB or less. (“Median=4096”) The average amount recovered was 56MB. (“Mean=56390”) 75% of the passes recovered 36MB or less. (“3rd Qu.=36860”) At least one pass recovered 2GB. (“Max.=2121000”) g1gc.df$Delta = g1gc.df$BeforeSize - g1gc.df$AfterSize summary(g1gc.df$Delta) ## Min. 1st Qu. Median Mean 3rd Qu. Max. ## 0 3070 4100 56400 36900 2120000 Q: What is the maximum User CPU time for a single collection? The worst garbage collection (“Max.”) is many standard deviations away from the mean. The data appears to be right skewed. summary(g1gc.df$UserTime) ## Min. 1st Qu. Median Mean 3rd Qu. Max. ## 0.040 0.470 0.620 0.751 0.920 3.370 sd(g1gc.df$UserTime) ## [1] 0.3966 Basic Graphics Once the data is in R, it is trivial to plot the data with formats including dot plots, line charts, bar charts (simple, stacked, grouped), pie charts, boxplots, scatter plots histograms, and kernel density plots. Histogram of User CPU Time per Collection I don't think that this graph requires any explanation. hist(g1gc.df$UserTime, main="User CPU Time per Collection", xlab="Seconds", ylab="Frequency") Box plot to identify outliers When the initial data is viewed with a box plot, you can see the one crazy outlier in the real time per GC. Save this data point for future analysis and drop the outlier so that it’s not throwing off our statistics. Now the box plot shows many outliers, which will be examined later, using times series analysis. Notice that the scale of the x-axis changes drastically once the crazy outlier is removed. par(mfrow=c(2,1)) boxplot(g1gc.df$UserTime,g1gc.df$SysTime,g1gc.df$RealTime, main="Box Plot of Time per GC\n(dominated by a crazy outlier)", names=c("usr","sys","elapsed"), xlab="Seconds per GC", ylab="Time (Seconds)", horizontal = TRUE, outcol="red") crazy.outlier.df=g1gc.df[g1gc.df$RealTime > 400,] g1gc.df=g1gc.df[g1gc.df$RealTime < 400,] boxplot(g1gc.df$UserTime,g1gc.df$SysTime,g1gc.df$RealTime, main="Box Plot of Time per GC\n(crazy outlier excluded)", names=c("usr","sys","elapsed"), xlab="Seconds per GC", ylab="Time (Seconds)", horizontal = TRUE, outcol="red") box(which = "outer", lty = "solid") Here is the crazy outlier for future analysis: crazy.outlier.df ## row.names SecondsSinceLaunch IncrementalCount ## 8233 2014-05-12T23:15:43.903-0700: 20741 8316 ## FullCount UserTime SysTime RealTime BeforeSize AfterSize TotalSize ## 8233 112 0.55 0.42 488.1 8381440 8235008 9437184 ## Delta ## 8233 146432 R Time Series Data To analyze the garbage collection as a time series, I’ll use Z’s Ordered Observations (zoo). “zoo is the creator for an S3 class of indexed totally ordered observations which includes irregular time series.” require(zoo) ## Loading required package: zoo ## ## Attaching package: 'zoo' ## ## The following objects are masked from 'package:base': ## ## as.Date, as.Date.numeric head(g1gc.df[,1]) ## [1] "2014-05-12T14:00:32.868-0700:" "2014-05-12T14:00:33.179-0700:" ## [3] "2014-05-12T14:00:33.677-0700:" "2014-05-12T14:00:35.538-0700:" ## [5] "2014-05-12T14:00:37.811-0700:" "2014-05-12T14:00:41.428-0700:" options("digits.secs"=3) times=as.POSIXct( g1gc.df[,1], format="%Y-%m-%dT%H:%M:%OS%z:") g1gc.z = zoo(g1gc.df[,-c(1)], order.by=times) head(g1gc.z) ## SecondsSinceLaunch IncrementalCount FullCount ## 2014-05-12 17:00:32.868 1.161 0 0 ## 2014-05-12 17:00:33.178 1.472 1 0 ## 2014-05-12 17:00:33.677 1.969 2 0 ## 2014-05-12 17:00:35.538 3.830 3 0 ## 2014-05-12 17:00:37.811 6.103 4 0 ## 2014-05-12 17:00:41.427 9.720 5 0 ## UserTime SysTime RealTime BeforeSize AfterSize ## 2014-05-12 17:00:32.868 0.11 0.04 0.02 8192 1400 ## 2014-05-12 17:00:33.178 0.05 0.01 0.02 5496 1672 ## 2014-05-12 17:00:33.677 0.04 0.01 0.01 5768 2557 ## 2014-05-12 17:00:35.538 0.21 0.05 0.04 22528 4907 ## 2014-05-12 17:00:37.811 0.08 0.01 0.02 24576 7072 ## 2014-05-12 17:00:41.427 0.26 0.06 0.04 43008 14336 ## TotalSize Delta ## 2014-05-12 17:00:32.868 9437184 6792 ## 2014-05-12 17:00:33.178 9437184 3824 ## 2014-05-12 17:00:33.677 9437184 3211 ## 2014-05-12 17:00:35.538 9437184 17621 ## 2014-05-12 17:00:37.811 9437184 17504 ## 2014-05-12 17:00:41.427 9437184 28672 Example of Two Benchmark Runs in One Log File The data in the following graph is from a different log file, not the one of primary interest to this article. I’m including this image because it is an example of idle periods followed by busy periods. It would be uninteresting to average the rate of garbage collection over the entire log file period. More interesting would be the rate of garbage collect in the two busy periods. Are they the same or different? Your production data may be similar, for example, bursts when employees return from lunch and idle times on weekend evenings, etc. Once the data is in an R Time Series, you can analyze isolated time windows. Clipping the Time Series data Flashing back to our test case… Viewing the data as a time series is interesting. You can see that the work intensive time period is between 9:00 PM and 3:00 AM. Lets clip the data to the interesting period:     par(mfrow=c(2,1)) plot(g1gc.z$UserTime, type="h", main="User Time per GC\nTime: Complete Log File", xlab="Time of Day", ylab="CPU Seconds per GC", col="#1b9e77") clipped.g1gc.z=window(g1gc.z, start=as.POSIXct("2014-05-12 21:00:00"), end=as.POSIXct("2014-05-13 03:00:00")) plot(clipped.g1gc.z$UserTime, type="h", main="User Time per GC\nTime: Limited to Benchmark Execution", xlab="Time of Day", ylab="CPU Seconds per GC", col="#1b9e77") box(which = "outer", lty = "solid") Cumulative Incremental and Full GC count Here is the cumulative incremental and full GC count. When the line is very steep, it indicates that the GCs are repeating very quickly. Notice that the scale on the Y axis is different for full vs. incremental. plot(clipped.g1gc.z[,c(2:3)], main="Cumulative Incremental and Full GC count", xlab="Time of Day", col="#1b9e77") GC Analysis of Benchmark Execution using Time Series data In the following series of 3 graphs: The “After Size” show the amount of heap space in use after each garbage collection. Many Java objects are still referenced, i.e. alive, during each garbage collection. This may indicate that the application has a memory leak, or may indicate that the application has a very large memory footprint. Typically, an application's memory footprint plateau's in the early stage of execution. One would expect this graph to have a flat top. The steep decline in the heap space may indicate that the application crashed after 2:00. The second graph shows that the outliers in real execution time, discussed above, occur near 2:00. when the Java heap seems to be quite full. The third graph shows that Full GCs are infrequent during the first few hours of execution. The rate of Full GC's, (the slope of the cummulative Full GC line), changes near midnight.   plot(clipped.g1gc.z[,c("AfterSize","RealTime","FullCount")], xlab="Time of Day", col=c("#1b9e77","red","#1b9e77")) GC Analysis of heap recovered Each GC trace includes the amount of heap space in use before and after the individual GC event. During garbage coolection, unreferenced objects are identified, the space holding the unreferenced objects is freed, and thus, the difference in before and after usage indicates how much space has been freed. The following box plot and bar chart both demonstrate the same point - the amount of heap space freed per garbage colloection is surprisingly low. par(mfrow=c(2,1)) boxplot(as.vector(clipped.g1gc.z$Delta), main="Amount of Heap Recovered per GC Pass", xlab="Size in KB", horizontal = TRUE, col="red") hist(as.vector(clipped.g1gc.z$Delta), main="Amount of Heap Recovered per GC Pass", xlab="Size in KB", breaks=100, col="red") box(which = "outer", lty = "solid") This graph is the most interesting. The dark blue area shows how much heap is occupied by referenced Java objects. This represents memory that holds live data. The red fringe at the top shows how much data was recovered after each garbage collection. barplot(clipped.g1gc.z[,c("AfterSize","Delta")], col=c("#7570b3","#e7298a"), xlab="Time of Day", border=NA) legend("topleft", c("Live Objects","Heap Recovered on GC"), fill=c("#7570b3","#e7298a")) box(which = "outer", lty = "solid") When I discuss the data in the log files with the customer, I will ask for an explaination for the large amount of referenced data resident in the Java heap. There are two are posibilities: There is a memory leak and the amount of space required to hold referenced objects will continue to grow, limited only by the maximum heap size. After the maximum heap size is reached, the JVM will throw an “Out of Memory” exception every time that the application tries to allocate a new object. If this is the case, the aplication needs to be debugged to identify why old objects are referenced when they are no longer needed. The application has a legitimate requirement to keep a large amount of data in memory. The customer may want to further increase the maximum heap size. Another possible solution would be to partition the application across multiple cluster nodes, where each node has responsibility for managing a unique subset of the data. Conclusion In conclusion, R is a very powerful tool for the analysis of Java garbage collection log files. The primary difficulty is data cleansing so that information can be read into an R data frame. Once the data has been read into R, a rich set of tools may be used for thorough evaluation.

    Read the article

  • Agile Development

    - by James Oloo Onyango
    Alot of literature has and is being written about agile developement and its surrounding philosophies. In my quest to find the best way to express the importance of agile methodologies, i have found Robert C. Martin's "A Satire Of Two Companies" to be both the most concise and thorough! Enjoy the read! Rufus Inc Project Kick Off Your name is Bob. The date is January 3, 2001, and your head still aches from the recent millennial revelry. You are sitting in a conference room with several managers and a group of your peers. You are a project team leader. Your boss is there, and he has brought along all of his team leaders. His boss called the meeting. "We have a new project to develop," says your boss's boss. Call him BB. The points in his hair are so long that they scrape the ceiling. Your boss's points are just starting to grow, but he eagerly awaits the day when he can leave Brylcream stains on the acoustic tiles. BB describes the essence of the new market they have identified and the product they want to develop to exploit this market. "We must have this new project up and working by fourth quarter October 1," BB demands. "Nothing is of higher priority, so we are cancelling your current project." The reaction in the room is stunned silence. Months of work are simply going to be thrown away. Slowly, a murmur of objection begins to circulate around the conference table.   His points give off an evil green glow as BB meets the eyes of everyone in the room. One by one, that insidious stare reduces each attendee to quivering lumps of protoplasm. It is clear that he will brook no discussion on this matter. Once silence has been restored, BB says, "We need to begin immediately. How long will it take you to do the analysis?" You raise your hand. Your boss tries to stop you, but his spitwad misses you and you are unaware of his efforts.   "Sir, we can't tell you how long the analysis will take until we have some requirements." "The requirements document won't be ready for 3 or 4 weeks," BB says, his points vibrating with frustration. "So, pretend that you have the requirements in front of you now. How long will you require for analysis?" No one breathes. Everyone looks around to see whether anyone has some idea. "If analysis goes beyond April 1, we have a problem. Can you finish the analysis by then?" Your boss visibly gathers his courage: "We'll find a way, sir!" His points grow 3 mm, and your headache increases by two Tylenol. "Good." BB smiles. "Now, how long will it take to do the design?" "Sir," you say. Your boss visibly pales. He is clearly worried that his 3 mms are at risk. "Without an analysis, it will not be possible to tell you how long design will take." BB's expression shifts beyond austere.   "PRETEND you have the analysis already!" he says, while fixing you with his vacant, beady little eyes. "How long will it take you to do the design?" Two Tylenol are not going to cut it. Your boss, in a desperate attempt to save his new growth, babbles: "Well, sir, with only six months left to complete the project, design had better take no longer than 3 months."   "I'm glad you agree, Smithers!" BB says, beaming. Your boss relaxes. He knows his points are secure. After a while, he starts lightly humming the Brylcream jingle. BB continues, "So, analysis will be complete by April 1, design will be complete by July 1, and that gives you 3 months to implement the project. This meeting is an example of how well our new consensus and empowerment policies are working. Now, get out there and start working. I'll expect to see TQM plans and QIT assignments on my desk by next week. Oh, and don't forget that your crossfunctional team meetings and reports will be needed for next month's quality audit." "Forget the Tylenol," you think to yourself as you return to your cubicle. "I need bourbon."   Visibly excited, your boss comes over to you and says, "Gosh, what a great meeting. I think we're really going to do some world shaking with this project." You nod in agreement, too disgusted to do anything else. "Oh," your boss continues, "I almost forgot." He hands you a 30-page document. "Remember that the SEI is coming to do an evaluation next week. This is the evaluation guide. You need to read through it, memorize it, and then shred it. It tells you how to answer any questions that the SEI auditors ask you. It also tells you what parts of the building you are allowed to take them to and what parts to avoid. We are determined to be a CMM level 3 organization by June!"   You and your peers start working on the analysis of the new project. This is difficult because you have no requirements. But from the 10-minute introduction given by BB on that fateful morning, you have some idea of what the product is supposed to do.   Corporate process demands that you begin by creating a use case document. You and your team begin enumerating use cases and drawing oval and stick diagrams. Philosophical debates break out among the team members. There is disagreement as to whether certain use cases should be connected with <<extends>> or <<includes>> relationships. Competing models are created, but nobody knows how to evaluate them. The debate continues, effectively paralyzing progress.   After a week, somebody finds the iceberg.com Web site, which recommends disposing entirely of <<extends>> and <<includes>> and replacing them with <<precedes>> and <<uses>>. The documents on this Web site, authored by Don Sengroiux, describes a method known as stalwart-analysis, which claims to be a step-by-step method for translating use cases into design diagrams. More competing use case models are created using this new scheme, but again, people can't agree on how to evaluate them. The thrashing continues. More and more, the use case meetings are driven by emotion rather than by reason. If it weren't for the fact that you don't have requirements, you'd be pretty upset by the lack of progress you are making. The requirements document arrives on February 15. And then again on February 20, 25, and every week thereafter. Each new version contradicts the previous one. Clearly, the marketing folks who are writing the requirements, empowered though they might be, are not finding consensus.   At the same time, several new competing use case templates have been proposed by the various team members. Each template presents its own particularly creative way of delaying progress. The debates rage on. On March 1, Prudence Putrigence, the process proctor, succeeds in integrating all the competing use case forms and templates into a single, all-encompassing form. Just the blank form is 15 pages long. She has managed to include every field that appeared on all the competing templates. She also presents a 159- page document describing how to fill out the use case form. All current use cases must be rewritten according to the new standard.   You marvel to yourself that it now requires 15 pages of fill-in-the-blank and essay questions to answer the question: What should the system do when the user presses Return? The corporate process (authored by L. E. Ott, famed author of "Holistic Analysis: A Progressive Dialectic for Software Engineers") insists that you discover all primary use cases, 87 percent of all secondary use cases, and 36.274 percent of all tertiary use cases before you can complete analysis and enter the design phase. You have no idea what a tertiary use case is. So in an attempt to meet this requirement, you try to get your use case document reviewed by the marketing department, which you hope will know what a tertiary use case is.   Unfortunately, the marketing folks are too busy with sales support to talk to you. Indeed, since the project started, you have not been able to get a single meeting with marketing, which has provided a never-ending stream of changing and contradictory requirements documents.   While one team has been spinning endlessly on the use case document, another team has been working out the domain model. Endless variations of UML documents are pouring out of this team. Every week, the model is reworked.   The team members can't decide whether to use <<interfaces>> or <<types>> in the model. A huge disagreement has been raging on the proper syntax and application of OCL. Others on the team just got back from a 5-day class on catabolism, and have been producing incredibly detailed and arcane diagrams that nobody else can fathom.   On March 27, with one week to go before analysis is to be complete, you have produced a sea of documents and diagrams but are no closer to a cogent analysis of the problem than you were on January 3. **** And then, a miracle happens.   **** On Saturday, April 1, you check your e-mail from home. You see a memo from your boss to BB. It states unequivocally that you are done with the analysis! You phone your boss and complain. "How could you have told BB that we were done with the analysis?" "Have you looked at a calendar lately?" he responds. "It's April 1!" The irony of that date does not escape you. "But we have so much more to think about. So much more to analyze! We haven't even decided whether to use <<extends>> or <<precedes>>!" "Where is your evidence that you are not done?" inquires your boss, impatiently. "Whaaa . . . ." But he cuts you off. "Analysis can go on forever; it has to be stopped at some point. And since this is the date it was scheduled to stop, it has been stopped. Now, on Monday, I want you to gather up all existing analysis materials and put them into a public folder. Release that folder to Prudence so that she can log it in the CM system by Monday afternoon. Then get busy and start designing."   As you hang up the phone, you begin to consider the benefits of keeping a bottle of bourbon in your bottom desk drawer. They threw a party to celebrate the on-time completion of the analysis phase. BB gave a colon-stirring speech on empowerment. And your boss, another 3 mm taller, congratulated his team on the incredible show of unity and teamwork. Finally, the CIO takes the stage to tell everyone that the SEI audit went very well and to thank everyone for studying and shredding the evaluation guides that were passed out. Level 3 now seems assured and will be awarded by June. (Scuttlebutt has it that managers at the level of BB and above are to receive significant bonuses once the SEI awards level 3.)   As the weeks flow by, you and your team work on the design of the system. Of course, you find that the analysis that the design is supposedly based on is flawedno, useless; no, worse than useless. But when you tell your boss that you need to go back and work some more on the analysis to shore up its weaker sections, he simply states, "The analysis phase is over. The only allowable activity is design. Now get back to it."   So, you and your team hack the design as best you can, unsure of whether the requirements have been properly analyzed. Of course, it really doesn't matter much, since the requirements document is still thrashing with weekly revisions, and the marketing department still refuses to meet with you.     The design is a nightmare. Your boss recently misread a book named The Finish Line in which the author, Mark DeThomaso, blithely suggested that design documents should be taken down to code-level detail. "If we are going to be working at that level of detail," you ask, "why don't we simply write the code instead?" "Because then you wouldn't be designing, of course. And the only allowable activity in the design phase is design!" "Besides," he continues, "we have just purchased a companywide license for Dandelion! This tool enables 'Round the Horn Engineering!' You are to transfer all design diagrams into this tool. It will automatically generate our code for us! It will also keep the design diagrams in sync with the code!" Your boss hands you a brightly colored shrinkwrapped box containing the Dandelion distribution. You accept it numbly and shuffle off to your cubicle. Twelve hours, eight crashes, one disk reformatting, and eight shots of 151 later, you finally have the tool installed on your server. You consider the week your team will lose while attending Dandelion training. Then you smile and think, "Any week I'm not here is a good week." Design diagram after design diagram is created by your team. Dandelion makes it very difficult to draw these diagrams. There are dozens and dozens of deeply nested dialog boxes with funny text fields and check boxes that must all be filled in correctly. And then there's the problem of moving classes between packages. At first, these diagram are driven from the use cases. But the requirements are changing so often that the use cases rapidly become meaningless. Debates rage about whether VISITOR or DECORATOR design patterns should be used. One developer refuses to use VISITOR in any form, claiming that it's not a properly object-oriented construct. Someone refuses to use multiple inheritance, since it is the spawn of the devil. Review meetings rapidly degenerate into debates about the meaning of object orientation, the definition of analysis versus design, or when to use aggregation versus association. Midway through the design cycle, the marketing folks announce that they have rethought the focus of the system. Their new requirements document is completely restructured. They have eliminated several major feature areas and replaced them with feature areas that they anticipate customer surveys will show to be more appropriate. You tell your boss that these changes mean that you need to reanalyze and redesign much of the system. But he says, "The analysis phase is system. But he says, "The analysis phase is over. The only allowable activity is design. Now get back to it."   You suggest that it might be better to create a simple prototype to show to the marketing folks and even some potential customers. But your boss says, "The analysis phase is over. The only allowable activity is design. Now get back to it." Hack, hack, hack, hack. You try to create some kind of a design document that might reflect the new requirements documents. However, the revolution of the requirements has not caused them to stop thrashing. Indeed, if anything, the wild oscillations of the requirements document have only increased in frequency and amplitude.   You slog your way through them.   On June 15, the Dandelion database gets corrupted. Apparently, the corruption has been progressive. Small errors in the DB accumulated over the months into bigger and bigger errors. Eventually, the CASE tool just stopped working. Of course, the slowly encroaching corruption is present on all the backups. Calls to the Dandelion technical support line go unanswered for several days. Finally, you receive a brief e-mail from Dandelion, informing you that this is a known problem and that the solution is to purchase the new version, which they promise will be ready some time next quarter, and then reenter all the diagrams by hand.   ****   Then, on July 1 another miracle happens! You are done with the design!   Rather than go to your boss and complain, you stock your middle desk drawer with some vodka.   **** They threw a party to celebrate the on-time completion of the design phase and their graduation to CMM level 3. This time, you find BB's speech so stirring that you have to use the restroom before it begins. New banners and plaques are all over your workplace. They show pictures of eagles and mountain climbers, and they talk about teamwork and empowerment. They read better after a few scotches. That reminds you that you need to clear out your file cabinet to make room for the brandy. You and your team begin to code. But you rapidly discover that the design is lacking in some significant areas. Actually, it's lacking any significance at all. You convene a design session in one of the conference rooms to try to work through some of the nastier problems. But your boss catches you at it and disbands the meeting, saying, "The design phase is over. The only allowable activity is coding. Now get back to it."   ****   The code generated by Dandelion is really hideous. It turns out that you and your team were using association and aggregation the wrong way, after all. All the generated code has to be edited to correct these flaws. Editing this code is extremely difficult because it has been instrumented with ugly comment blocks that have special syntax that Dandelion needs in order to keep the diagrams in sync with the code. If you accidentally alter one of these comments, the diagrams will be regenerated incorrectly. It turns out that "Round the Horn Engineering" requires an awful lot of effort. The more you try to keep the code compatible with Dandelion, the more errors Dandelion generates. In the end, you give up and decide to keep the diagrams up to date manually. A second later, you decide that there's no point in keeping the diagrams up to date at all. Besides, who has time?   Your boss hires a consultant to build tools to count the number of lines of code that are being produced. He puts a big thermometer graph on the wall with the number 1,000,000 on the top. Every day, he extends the red line to show how many lines have been added. Three days after the thermometer appears on the wall, your boss stops you in the hall. "That graph isn't growing quickly enough. We need to have a million lines done by October 1." "We aren't even sh-sh-sure that the proshect will require a m-million linezh," you blather. "We have to have a million lines done by October 1," your boss reiterates. His points have grown again, and the Grecian formula he uses on them creates an aura of authority and competence. "Are you sure your comment blocks are big enough?" Then, in a flash of managerial insight, he says, "I have it! I want you to institute a new policy among the engineers. No line of code is to be longer than 20 characters. Any such line must be split into two or more preferably more. All existing code needs to be reworked to this standard. That'll get our line count up!"   You decide not to tell him that this will require two unscheduled work months. You decide not to tell him anything at all. You decide that intravenous injections of pure ethanol are the only solution. You make the appropriate arrangements. Hack, hack, hack, and hack. You and your team madly code away. By August 1, your boss, frowning at the thermometer on the wall, institutes a mandatory 50-hour workweek.   Hack, hack, hack, and hack. By September 1st, the thermometer is at 1.2 million lines and your boss asks you to write a report describing why you exceeded the coding budget by 20 percent. He institutes mandatory Saturdays and demands that the project be brought back down to a million lines. You start a campaign of remerging lines. Hack, hack, hack, and hack. Tempers are flaring; people are quitting; QA is raining trouble reports down on you. Customers are demanding installation and user manuals; salespeople are demanding advance demonstrations for special customers; the requirements document is still thrashing, the marketing folks are complaining that the product isn't anything like they specified, and the liquor store won't accept your credit card anymore. Something has to give.    On September 15, BB calls a meeting. As he enters the room, his points are emitting clouds of steam. When he speaks, the bass overtones of his carefully manicured voice cause the pit of your stomach to roll over. "The QA manager has told me that this project has less than 50 percent of the required features implemented. He has also informed me that the system crashes all the time, yields wrong results, and is hideously slow. He has also complained that he cannot keep up with the continuous train of daily releases, each more buggy than the last!" He stops for a few seconds, visibly trying to compose himself. "The QA manager estimates that, at this rate of development, we won't be able to ship the product until December!" Actually, you think it's more like March, but you don't say anything. "December!" BB roars with such derision that people duck their heads as though he were pointing an assault rifle at them. "December is absolutely out of the question. Team leaders, I want new estimates on my desk in the morning. I am hereby mandating 65-hour work weeks until this project is complete. And it better be complete by November 1."   As he leaves the conference room, he is heard to mutter: "Empowermentbah!" * * * Your boss is bald; his points are mounted on BB's wall. The fluorescent lights reflecting off his pate momentarily dazzle you. "Do you have anything to drink?" he asks. Having just finished your last bottle of Boone's Farm, you pull a bottle of Thunderbird from your bookshelf and pour it into his coffee mug. "What's it going to take to get this project done? " he asks. "We need to freeze the requirements, analyze them, design them, and then implement them," you say callously. "By November 1?" your boss exclaims incredulously. "No way! Just get back to coding the damned thing." He storms out, scratching his vacant head.   A few days later, you find that your boss has been transferred to the corporate research division. Turnover has skyrocketed. Customers, informed at the last minute that their orders cannot be fulfilled on time, have begun to cancel their orders. Marketing is re-evaluating whether this product aligns with the overall goals of the company. Memos fly, heads roll, policies change, and things are, overall, pretty grim. Finally, by March, after far too many sixty-five hour weeks, a very shaky version of the software is ready. In the field, bug-discovery rates are high, and the technical support staff are at their wits' end, trying to cope with the complaints and demands of the irate customers. Nobody is happy.   In April, BB decides to buy his way out of the problem by licensing a product produced by Rupert Industries and redistributing it. The customers are mollified, the marketing folks are smug, and you are laid off.     Rupert Industries: Project Alpha   Your name is Robert. The date is January 3, 2001. The quiet hours spent with your family this holiday have left you refreshed and ready for work. You are sitting in a conference room with your team of professionals. The manager of the division called the meeting. "We have some ideas for a new project," says the division manager. Call him Russ. He is a high-strung British chap with more energy than a fusion reactor. He is ambitious and driven but understands the value of a team. Russ describes the essence of the new market opportunity the company has identified and introduces you to Jane, the marketing manager, who is responsible for defining the products that will address it. Addressing you, Jane says, "We'd like to start defining our first product offering as soon as possible. When can you and your team meet with me?" You reply, "We'll be done with the current iteration of our project this Friday. We can spare a few hours for you between now and then. After that, we'll take a few people from the team and dedicate them to you. We'll begin hiring their replacements and the new people for your team immediately." "Great," says Russ, "but I want you to understand that it is critical that we have something to exhibit at the trade show coming up this July. If we can't be there with something significant, we'll lose the opportunity."   "I understand," you reply. "I don't yet know what it is that you have in mind, but I'm sure we can have something by July. I just can't tell you what that something will be right now. In any case, you and Jane are going to have complete control over what we developers do, so you can rest assured that by July, you'll have the most important things that can be accomplished in that time ready to exhibit."   Russ nods in satisfaction. He knows how this works. Your team has always kept him advised and allowed him to steer their development. He has the utmost confidence that your team will work on the most important things first and will produce a high-quality product.   * * *   "So, Robert," says Jane at their first meeting, "How does your team feel about being split up?" "We'll miss working with each other," you answer, "but some of us were getting pretty tired of that last project and are looking forward to a change. So, what are you people cooking up?" Jane beams. "You know how much trouble our customers currently have . . ." And she spends a half hour or so describing the problem and possible solution. "OK, wait a second" you respond. "I need to be clear about this." And so you and Jane talk about how this system might work. Some of her ideas aren't fully formed. You suggest possible solutions. She likes some of them. You continue discussing.   During the discussion, as each new topic is addressed, Jane writes user story cards. Each card represents something that the new system has to do. The cards accumulate on the table and are spread out in front of you. Both you and Jane point at them, pick them up, and make notes on them as you discuss the stories. The cards are powerful mnemonic devices that you can use to represent complex ideas that are barely formed.   At the end of the meeting, you say, "OK, I've got a general idea of what you want. I'm going to talk to the team about it. I imagine they'll want to run some experiments with various database structures and presentation formats. Next time we meet, it'll be as a group, and we'll start identifying the most important features of the system."   A week later, your nascent team meets with Jane. They spread the existing user story cards out on the table and begin to get into some of the details of the system. The meeting is very dynamic. Jane presents the stories in the order of their importance. There is much discussion about each one. The developers are concerned about keeping the stories small enough to estimate and test. So they continually ask Jane to split one story into several smaller stories. Jane is concerned that each story have a clear business value and priority, so as she splits them, she makes sure that this stays true.   The stories accumulate on the table. Jane writes them, but the developers make notes on them as needed. Nobody tries to capture everything that is said; the cards are not meant to capture everything but are simply reminders of the conversation.   As the developers become more comfortable with the stories, they begin writing estimates on them. These estimates are crude and budgetary, but they give Jane an idea of what the story will cost.   At the end of the meeting, it is clear that many more stories could be discussed. It is also clear that the most important stories have been addressed and that they represent several months worth of work. Jane closes the meeting by taking the cards with her and promising to have a proposal for the first release in the morning.   * * *   The next morning, you reconvene the meeting. Jane chooses five cards and places them on the table. "According to your estimates, these cards represent about one perfect team-week's worth of work. The last iteration of the previous project managed to get one perfect team-week done in 3 real weeks. If we can get these five stories done in 3 weeks, we'll be able to demonstrate them to Russ. That will make him feel very comfortable about our progress." Jane is pushing it. The sheepish look on her face lets you know that she knows it too. You reply, "Jane, this is a new team, working on a new project. It's a bit presumptuous to expect that our velocity will be the same as the previous team's. However, I met with the team yesterday afternoon, and we all agreed that our initial velocity should, in fact, be set to one perfectweek for every 3 real-weeks. So you've lucked out on this one." "Just remember," you continue, "that the story estimates and the story velocity are very tentative at this point. We'll learn more when we plan the iteration and even more when we implement it."   Jane looks over her glasses at you as if to say "Who's the boss around here, anyway?" and then smiles and says, "Yeah, don't worry. I know the drill by now."Jane then puts 15 more cards on the table. She says, "If we can get all these cards done by the end of March, we can turn the system over to our beta test customers. And we'll get good feedback from them."   You reply, "OK, so we've got our first iteration defined, and we have the stories for the next three iterations after that. These four iterations will make our first release."   "So," says Jane, can you really do these five stories in the next 3 weeks?" "I don't know for sure, Jane," you reply. "Let's break them down into tasks and see what we get."   So Jane, you, and your team spend the next several hours taking each of the five stories that Jane chose for the first iteration and breaking them down into small tasks. The developers quickly realize that some of the tasks can be shared between stories and that other tasks have commonalities that can probably be taken advantage of. It is clear that potential designs are popping into the developers' heads. From time to time, they form little discussion knots and scribble UML diagrams on some cards.   Soon, the whiteboard is filled with the tasks that, once completed, will implement the five stories for this iteration. You start the sign-up process by saying, "OK, let's sign up for these tasks." "I'll take the initial database generation." Says Pete. "That's what I did on the last project, and this doesn't look very different. I estimate it at two of my perfect workdays." "OK, well, then, I'll take the login screen," says Joe. "Aw, darn," says Elaine, the junior member of the team, "I've never done a GUI, and kinda wanted to try that one."   "Ah, the impatience of youth," Joe says sagely, with a wink in your direction. "You can assist me with it, young Jedi." To Jane: "I think it'll take me about three of my perfect workdays."   One by one, the developers sign up for tasks and estimate them in terms of their own perfect workdays. Both you and Jane know that it is best to let the developers volunteer for tasks than to assign the tasks to them. You also know full well that you daren't challenge any of the developers' estimates. You know these people, and you trust them. You know that they are going to do the very best they can.   The developers know that they can't sign up for more perfect workdays than they finished in the last iteration they worked on. Once each developer has filled his or her schedule for the iteration, they stop signing up for tasks.   Eventually, all the developers have stopped signing up for tasks. But, of course, tasks are still left on the board.   "I was worried that that might happen," you say, "OK, there's only one thing to do, Jane. We've got too much to do in this iteration. What stories or tasks can we remove?" Jane sighs. She knows that this is the only option. Working overtime at the beginning of a project is insane, and projects where she's tried it have not fared well.   So Jane starts to remove the least-important functionality. "Well, we really don't need the login screen just yet. We can simply start the system in the logged-in state." "Rats!" cries Elaine. "I really wanted to do that." "Patience, grasshopper." says Joe. "Those who wait for the bees to leave the hive will not have lips too swollen to relish the honey." Elaine looks confused. Everyone looks confused. "So . . .," Jane continues, "I think we can also do away with . . ." And so, bit by bit, the list of tasks shrinks. Developers who lose a task sign up for one of the remaining ones.   The negotiation is not painless. Several times, Jane exhibits obvious frustration and impatience. Once, when tensions are especially high, Elaine volunteers, "I'll work extra hard to make up some of the missing time." You are about to correct her when, fortunately, Joe looks her in the eye and says, "When once you proceed down the dark path, forever will it dominate your destiny."   In the end, an iteration acceptable to Jane is reached. It's not what Jane wanted. Indeed, it is significantly less. But it's something the team feels that can be achieved in the next 3 weeks.   And, after all, it still addresses the most important things that Jane wanted in the iteration. "So, Jane," you say when things had quieted down a bit, "when can we expect acceptance tests from you?" Jane sighs. This is the other side of the coin. For every story the development team implements,   Jane must supply a suite of acceptance tests that prove that it works. And the team needs these long before the end of the iteration, since they will certainly point out differences in the way Jane and the developers imagine the system's behaviour.   "I'll get you some example test scripts today," Jane promises. "I'll add to them every day after that. You'll have the entire suite by the middle of the iteration."   * * *   The iteration begins on Monday morning with a flurry of Class, Responsibilities, Collaborators sessions. By midmorning, all the developers have assembled into pairs and are rapidly coding away. "And now, my young apprentice," Joe says to Elaine, "you shall learn the mysteries of test-first design!"   "Wow, that sounds pretty rad," Elaine replies. "How do you do it?" Joe beams. It's clear that he has been anticipating this moment. "OK, what does the code do right now?" "Huh?" replied Elaine, "It doesn't do anything at all; there is no code."   "So, consider our task; can you think of something the code should do?" "Sure," Elaine said with youthful assurance, "First, it should connect to the database." "And thereupon, what must needs be required to connecteth the database?" "You sure talk weird," laughed Elaine. "I think we'd have to get the database object from some registry and call the Connect() method. "Ah, astute young wizard. Thou perceives correctly that we requireth an object within which we can cacheth the database object." "Is 'cacheth' really a word?" "It is when I say it! So, what test can we write that we know the database registry should pass?" Elaine sighs. She knows she'll just have to play along. "We should be able to create a database object and pass it to the registry in a Store() method. And then we should be able to pull it out of the registry with a Get() method and make sure it's the same object." "Oh, well said, my prepubescent sprite!" "Hay!" "So, now, let's write a test function that proves your case." "But shouldn't we write the database object and registry object first?" "Ah, you've much to learn, my young impatient one. Just write the test first." "But it won't even compile!" "Are you sure? What if it did?" "Uh . . ." "Just write the test, Elaine. Trust me." And so Joe, Elaine, and all the other developers began to code their tasks, one test case at a time. The room in which they worked was abuzz with the conversations between the pairs. The murmur was punctuated by an occasional high five when a pair managed to finish a task or a difficult test case.   As development proceeded, the developers changed partners once or twice a day. Each developer got to see what all the others were doing, and so knowledge of the code spread generally throughout the team.   Whenever a pair finished something significant whether a whole task or simply an important part of a task they integrated what they had with the rest of the system. Thus, the code base grew daily, and integration difficulties were minimized.   The developers communicated with Jane on a daily basis. They'd go to her whenever they had a question about the functionality of the system or the interpretation of an acceptance test case.   Jane, good as her word, supplied the team with a steady stream of acceptance test scripts. The team read these carefully and thereby gained a much better understanding of what Jane expected the system to do. By the beginning of the second week, there was enough functionality to demonstrate to Jane. She watched eagerly as the demonstration passed test case after test case. "This is really cool," Jane said as the demonstration finally ended. "But this doesn't seem like one-third of the tasks. Is your velocity slower than anticipated?"   You grimace. You'd been waiting for a good time to mention this to Jane but now she was forcing the issue. "Yes, unfortunately, we are going more slowly than we had expected. The new application server we are using is turning out to be a pain to configure. Also, it takes forever to reboot, and we have to reboot it whenever we make even the slightest change to its configuration."   Jane eyes you with suspicion. The stress of last Monday's negotiations had still not entirely dissipated. She says, "And what does this mean to our schedule? We can't slip it again, we just can't. Russ will have a fit! He'll haul us all into the woodshed and ream us some new ones."   You look Jane right in the eyes. There's no pleasant way to give someone news like this. So you just blurt out, "Look, if things keep going like they're going, we're not going to be done with everything by next Friday. Now it's possible that we'll figure out a way to go faster. But, frankly, I wouldn't depend on that. You should start thinking about one or two tasks that could be removed from the iteration without ruining the demonstration for Russ. Come hell or high water, we are going to give that demonstration on Friday, and I don't think you want us to choose which tasks to omit."   "Aw forchrisakes!" Jane barely manages to stifle yelling that last word as she stalks away, shaking her head. Not for the first time, you say to yourself, "Nobody ever promised me project management would be easy." You are pretty sure it won't be the last time, either.   Actually, things went a bit better than you had hoped. The team did, in fact, have to drop one task from the iteration, but Jane had chosen wisely, and the demonstration for Russ went without a hitch. Russ was not impressed with the progress, but neither was he dismayed. He simply said, "This is pretty good. But remember, we have to be able to demonstrate this system at the trade show in July, and at this rate, it doesn't look like you'll have all that much to show." Jane, whose attitude had improved dramatically with the completion of the iteration, responded to Russ by saying, "Russ, this team is working hard, and well. When July comes around, I am confident that we'll have something significant to demonstrate. It won't be everything, and some of it may be smoke and mirrors, but we'll have something."   Painful though the last iteration was, it had calibrated your velocity numbers. The next iteration went much better. Not because your team got more done than in the last iteration but simply because the team didn't have to remove any tasks or stories in the middle of the iteration.   By the start of the fourth iteration, a natural rhythm has been established. Jane, you, and the team know exactly what to expect from one another. The team is running hard, but the pace is sustainable. You are confident that the team can keep up this pace for a year or more.   The number of surprises in the schedule diminishes to near zero; however, the number of surprises in the requirements does not. Jane and Russ frequently look over the growing system and make recommendations or changes to the existing functionality. But all parties realize that these changes take time and must be scheduled. So the changes do not cause anyone's expectations to be violated. In March, there is a major demonstration of the system to the board of directors. The system is very limited and is not yet in a form good enough to take to the trade show, but progress is steady, and the board is reasonably impressed.   The second release goes even more smoothly than the first. By now, the team has figured out a way to automate Jane's acceptance test scripts. The team has also refactored the design of the system to the point that it is really easy to add new features and change old ones. The second release was done by the end of June and was taken to the trade show. It had less in it than Jane and Russ would have liked, but it did demonstrate the most important features of the system. Although customers at the trade show noticed that certain features were missing, they were very impressed overall. You, Russ, and Jane all returned from the trade show with smiles on your faces. You all felt as though this project was a winner.   Indeed, many months later, you are contacted by Rufus Inc. That company had been working on a system like this for its internal operations. Rufus has canceled the development of that system after a death-march project and is negotiating to license your technology for its environment.   Indeed, things are looking up!

    Read the article

< Previous Page | 145 146 147 148 149