- logback-user - mailman3.qos.ch

Class not found exception for LoggingEventVO
by MDerr＠nycm.com 13 Jan '12

13 Jan '12

I have an application client set up to send a logging message over a bus using a logback JMSQueueAppender. Logback version is 0.9.24 and slf4j version is 1.6.1. This is running on a Websphere 7.0 server in Rational Application Developer 7.5. I've tested the bus and queue with a simple text message using the RAD Universal Test Client and that works fine, so at least the foundation is solid. When I log a message to the message queue, I wind up with the following stack trace under a wide variety of attempts: [8/11/10 14:44:16:223 EDT] 0000002b SystemOut O Exception CWSIA0122E: An exception occurred deserializing a message, exception: java.lang.ClassNotFoundException: ch.qos.logback.classic.spi.LoggingEventVO. [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R javax.jms.MessageFormatException: CWSIA0122E: An exception occurred deserializing a message, exception: java.lang.ClassNotFoundException: ch.qos.logback.classic.spi.LoggingEventVO. [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.sib.api.jms.impl.JmsObjectMessageImpl.getObject( JmsObjectMessageImpl.java:277) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.csg.logger.mdb.NYCMLoggerMDB.handleMessage(NYCMLoggerMDB.java:45) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.csg.logger.mdb.NYCMLoggerMDB.onMessage(NYCMLoggerMDB.java:36) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ejs.container.MessageEndpointHandler.invokeMdbMethod( MessageEndpointHandler.java:1092) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ejs.container.MessageEndpointHandler.invoke( MessageEndpointHandler.java:777) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at $Proxy42.onMessage(Unknown Source) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.sib.api.jmsra.impl.JmsJcaEndpointInvokerImpl.invokeEndpoint( JmsJcaEndpointInvokerImpl.java:233) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.sib.ra.inbound.impl.SibRaDispatcher.dispatch( SibRaDispatcher.java:900) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.sib.ra.inbound.impl.SibRaSingleProcessListener$SibRaWork.run( SibRaSingleProcessListener.java:550) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ejs.j2c.work.WorkProxy.run(WorkProxy.java:399) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.util.ThreadPool$Worker.run(ThreadPool.java:1550) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R Caused by: java.lang.ClassNotFoundException: ch.qos.logback.classic.spi.LoggingEventVO [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at java.lang.Class.forNameImpl(Native Method) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at java.lang.Class.forName(Class.java:136) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.util.WsObjectInputStream.loadClass(WsObjectInputStream.java:227 ) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.util.WsObjectInputStream.access$000(WsObjectInputStream.java:53 ) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.util.WsObjectInputStream$2.run(WsObjectInputStream.java:191) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.security.util.AccessController.doPrivileged( AccessController.java:118) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.util.WsObjectInputStream.resolveClass( WsObjectInputStream.java:164) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at com.ibm.ws.util.WsObjectInputStream.resolveClass( WsObjectInputStream.java:156) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at java.io.ObjectInputStream.readNonProxyDesc(ObjectInputStream.java:1591) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at java.io.ObjectInputStream.readClassDesc(ObjectInputStream.java:1512) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:1748) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1345) [8/11/10 14:44:16:223 EDT] 0000002b SystemErr R at java.io.ObjectInputStream.readObject(ObjectInputStream.java:364) [8/11/10 14:44:16:239 EDT] 0000002b SystemErr R at com.ibm.ws.sib.api.jms.impl.JmsObjectMessageImpl.getObjectInternal( JmsObjectMessageImpl.java:345) [8/11/10 14:44:16:239 EDT] 0000002b SystemErr R at com.ibm.ws.sib.api.jms.impl.JmsObjectMessageImpl.getObject( JmsObjectMessageImpl.java:271) [8/11/10 14:44:16:239 EDT] 0000002b SystemErr R ... 10 more This always happens when trying to get the message object out of the message in the onMessage code in my message driven bean. For example, the stack dump was generated at the event = . . . line in the code below: if (msg instanceof ObjectMessage){ ObjectMessage objectMesssage = (ObjectMessage) msg; System.out.println("Got ObjectMessage"); event = (ILoggingEvent) objectMesssage.getObject(); } I have the logback classic jar on my classpath successfully because I use a number of other classes from it in my code. If anyone can help with this, I would appreciate it. I can provide additional information as needed. Thanks, Merritt Derr ***CONFIDENTIALITY NOTICE*** This email and any attachments to it are confidential and intended solely for the individual or entity to whom it is addressed. Any unauthorized review, use, disclosure or distribution is prohibited. If you have received this email in error, please contact the sender by reply email and destroy all copies of the original message.

3 2

Object rendering
by Manuel Uberti 13 Jan '12

13 Jan '12

Hi, I am new to logback, and I was asked to migrate our project from log4j to logback. I was wondering if there's any kind of support for object rendering such as in log4j. This is what we're now doing with log4j. public static LoggerRepository getLoggerRepository() { if (loggerRepository == null) { loggerRepository = org.apache.log4j.LogManager.getLoggerRepository(); } return loggerRepository; } public static RendererSupport getRendererSupport() { LoggerRepository lr = getLoggerRepository(); if ((lr != null) && (lr instanceof RendererSupport)) rendererSupport = (RendererSupport) lr; return rendererSupport; } public static ObjectRender getRendererObject(Class<?> clazz) { ObjectRender renderer = getRendererSupport().getRendererMap().get(clazz); return renderer; } Any thoughts? Thank you, Manuel

1 0

Logging separation with JavaSE
by BRESCIANI, MARCO (MARCO) 13 Jan '12

13 Jan '12

Hello all, I've read the logging separation chapter of the manual since my customers (and our tech support) are requiring to separate log file for each "application". Brief description of the application: something like, say, Word: a container application (with a MDC called %X{Server} with a string, say, "Main") plus some different "internal" applications loaded at runtime (with a MDC called %X{Address} being an IP address). When the application starts there is the %X{Server} MDC while when on of the internal applications is loaded by the main one, such application sets the MDC with its own IP. Seems working fine. :) I've been asked to split the log files according to the MDC (because it's "too difficult to use a regexp to filter the file". You know, customers...) and reading the logging separation chapter, I modified my appender: <appender class="ch.qos.logback.core.rolling.RollingFileAppender" name="RootFileAppender" encoding="UTF-8"> <file>trace/WTPMS.log</file> Into this: <appender class="ch.qos.logback.core.rolling.RollingFileAppender" name="RootFileAppender" encoding="UTF-8"> <file>trace/${Server}.log</file> And logback creates a file called "Server_IS_UNDEFINED.log", probably because the MDC is set after few (4-5) lines of checks that include a logging entry. Also, setting the MDC, anyway does not create a new file with the new name, even if the "scan" property is true. Any possible solution to this kind of separation? Thanks all, MARCO BRESCIANI ALCATEL-LUCENT Marco.Bresciani(a)alcatel-lucent.com

1 0

SMTPAppender for android
by Enrico Spinielli 12 Jan '12

12 Jan '12

Hi, I am just trying out logback for Android and I was curious about the possibility to use SMTPAppender, but found out it is not included in the features set. Is there any technical reason why this appender is not supported? Or is it just because of "logistics" (aka lack of available effort/time ;-)? Thanks a lot for your feedback Bye -- Enrico Spinielli "Do Androids dream of electric sheep?"— Philip K. Dick "Hear and forget; see and remember;do and understand."—Mitchel Resnick "He who refuses to do arithmetic is doomed to talk nonsense."—John McCarthy

2 2

appender file name using name of warfile directory
by Volker Hochstein 11 Jan '12

11 Jan '12

Hi, just a quick question. let s assume I ve got a warfile application1 installed to tomcats webapps dir. Do I have the possiblilty to do the following in my appender, without defining any properties of my own? <file>${catalina.base}/logs/${somehow find out that it is application1 running}.log Basically, if there is a property like catalina.base, which gives me the of the dir in webapps directory? Thanks a lot in adance. -- Volker

1 0

How to disable PackagingDataCalculator
by Christian Wurbs 09 Jan '12

09 Jan '12

1 0

Re: [logback-user] JaninoEventEvaluator and class loader
by Glyn Normington 09 Jan '12

09 Jan '12

Arno kindly fixed the Janino mailing list archive, as promised, so here's the link to my thread there: [1] http://old.nabble.com/Janino-and-OSGi--to33030424.html Regards, Glyn On 8 Jan 2012, at 18:50, Arno Unkrig wrote: > Hi Glyn and everybody else, > > one quick note here: There is no difference between "ScriptEvaluator" and "ExpressionEvaluator" (other than that the former compiles a method body and the latter an expression). Initially, both came with a set of constructors with varying parameter counts and types, but because the configuration possibilities grew over time, the recommended usage pattern is now to call the parameterless constructor, then a series of setters, and then (one of) the "cook()" methods. > > I also wonder why LOGBACK switched from ExpressionEvaluator > > https://github.com/ceki/logback/blob/6c5ba501831d19879e6865f795a1c294ad25bf… > > to ScriptEvaluator > > https://github.com/ceki/logback/blob/06a5b692f14560636bd92d7bd7cf1f85830f4e… > > anyway!? That's a huge semantical change... > > Am 05.01.2012 13:09, schrieb Glyn Normington: >> I am attempting to get Janino 2.6.1 working with Logback 0.9.28 (or >> later, but that's the version we are using in Eclipse Virgo right now) - >> see [0] for background. Unfortunately Janino ends up using the thread >> context class loader as its "parent" class loader and fails with a >> runtime exception. I have discussed this ([1]) on the Janino mailing >> list and it seems that it is necessary to set the parent class loader in >> Janino. I am somewhat at the mercy of Logback here. >> >> The surprising thing is that the code in Logback 0.9.24 looked pretty >> usable in this respect. JaninoEventEvaluator.start had the following >> sequence: >> >> ClassLoader cl = context.getClass().getClassLoader(); >> ee = new ExpressionEvaluator(getDecoratedExpression(), EXPRESSION_TYPE, >> getParameterNames(), getParameterTypes(), THROWN_EXCEPTIONS, cl); >> >> thus setting the parent class loader to a value which I could ensure >> would be capable of loading the necessary types. >> >> In 0.9.28 this code has been replaced by: >> >> scriptEvaluator = new ScriptEvaluator(getDecoratedExpression(), >> EXPRESSION_TYPE, >> getParameterNames(), getParameterTypes(), THROWN_EXCEPTIONS); >> >> which causes the current TCCL to be used as the parent class loader and >> ultimately results in Janino failing. >> >> I can't control the TCCL that happens to be in use when start is called >> as that is driven out of a logging call which can come from an arbitrary >> thread. >> >> I found the relevant commit ([2]), but I can't tell from that why this >> specific change was made. >> >> If it is absolutely necessary to use ScriptEvaluator rather than >> ExpressionEvaluator, the following code sequence (based on a suggestion >> from Arno Unkrig) would reproduce the parent class loader behaviour of >> 0.9.24: >> >> ClassLoader cl = context.getClass().getClassLoader(); >> scriptEvaluator = new ScriptEvaluator(); >> scriptEvaluator.setParentClassLoader(cl); >> scriptEvaluator.setReturnType(EXPRESSION_TYPE); >> scriptEvaluator.setParameterNames(getParameterNames()); >> scriptEvaluator.setParameterTypes(getParameterTypes()); >> scriptEvaluator.setThrownExceptions(THROWN_EXCEPTIONS); >> scriptEvaluator.cook(getDecoratedExpression()); >> >> The alternative of using ExpressionEvaluator is much neater and seems to >> be close to the ScriptEvaluator variant since ExpressionEvaluator >> extends ScriptEvaluator. >> >> Any suggestions gratefully received. >> >> Regards, >> Glyn >> [0] https://bugs.eclipse.org/bugs/show_bug.cgi?id=333920 >> [1] mailing list archive link currently broken - if I can get a link to >> it, I'll post this later >> [2] >> https://github.com/ceki/logback/commit/06a5b692f14560636bd92d7bd7cf1f85830f… >> >

1 1

Log Messages duplicate when running unit tests.
by Matt Bertolini 06 Jan '12

06 Jan '12

Hello. I am trying to implement logback/slf4j and an extremely frustrating thing is occurring when I run my unit tests. All of the log messages are being duplicated. I am aware of the whole additive nature of appenders and I am pretty sure that I am doing things correctly in that regard but the messages continue to duplicate when running unit tests. When I run through a main method, they do not. Here is my configuration and the messages that are displayed when running unit tests. Dependencies I have included (via Ivy): <dependency org="org.slf4j" name="slf4j-api" rev="1.6.4"/> <dependency org="ch.qos.logback" name="logback-core" rev="1.0.0"/> <dependency org="ch.qos.logback" name="logback-classic" rev="1.0.0"/> Configuration: <configuration debug="true"> <contextName>example-test</contextName> <appender name="STDOUT" class="ch.qos.logback.core.ConsoleAppender"> <encoder> <pattern>%d{HH:mm:ss.SSS} %-5level %logger - %msg%n</pattern> </encoder> </appender> <logger name="com.example" level="debug"/> <root level="warn"> <appender-ref ref="STDOUT"/> </root> </configuration> Log output when running unit tests: 13:44:23,308 |-INFO in ch.qos.logback.classic.joran.action.ContextNameAction - Setting logger context name as [example-test] 13:44:23,308 |-INFO in ch.qos.logback.core.joran.action.AppenderAction - About to instantiate appender of type [ch.qos.logback.core.ConsoleAppender] 13:44:23,313 |-INFO in ch.qos.logback.core.joran.action.AppenderAction - Naming appender as [STDOUT] 13:44:23,344 |-INFO in ch.qos.logback.core.joran.action.NestedComplexPropertyIA - Assuming default type [ch.qos.logback.classic.encoder.PatternLayoutEncoder] for [encoder] property 13:44:23,431 |-INFO in ch.qos.logback.classic.joran.action.LoggerAction - Setting level of logger [com.example] to DEBUG 13:44:23,431 |-INFO in ch.qos.logback.classic.joran.action.RootLoggerAction - Setting level of ROOT logger to WARN 13:44:23,431 |-INFO in ch.qos.logback.core.joran.action.AppenderRefAction - Attaching appender named [STDOUT] to Logger[ROOT] 13:44:23,433 |-INFO in ch.qos.logback.classic.joran.action.ConfigurationAction - End of configuration. 13:44:23.453 INFO com.example.project.service.impl.ExampleClassImpl - This is a test. 13:44:23.466 INFO com.example.project.service.impl.ExampleClassImpl - This is a test. Log output when running under main method: 13:48:16,729 |-INFO in ch.qos.logback.classic.joran.action.ContextNameAction - Setting logger context name as [example-test] 13:48:16,729 |-INFO in ch.qos.logback.core.joran.action.AppenderAction - About to instantiate appender of type [ch.qos.logback.core.ConsoleAppender] 13:48:16,734 |-INFO in ch.qos.logback.core.joran.action.AppenderAction - Naming appender as [STDOUT] 13:48:16,760 |-INFO in ch.qos.logback.core.joran.action.NestedComplexPropertyIA - Assuming default type [ch.qos.logback.classic.encoder.PatternLayoutEncoder] for [encoder] property 13:48:16,825 |-INFO in ch.qos.logback.classic.joran.action.LoggerAction - Setting level of logger [com.example] to DEBUG 13:48:16,825 |-INFO in ch.qos.logback.classic.joran.action.RootLoggerAction - Setting level of ROOT logger to WARN 13:48:16,825 |-INFO in ch.qos.logback.core.joran.action.AppenderRefAction - Attaching appender named [STDOUT] to Logger[ROOT] 13:48:16,827 |-INFO in ch.qos.logback.classic.joran.action.ConfigurationAction - End of configuration. 13:48:16.843 INFO com.example.project.service.impl.ExampleClassImpl - This is a test. I know that the unit tests are picking up my configuration correctly. When I examine my unit test results, I can see that tests are only being run once so I it is not a duplicate execution of the code. Is there anything special I have to do to stop the duplicates when running unit tests? Thanks for your help. --Matt--

2 2

JaninoEventEvaluator and class loader
by Glyn Normington 06 Jan '12

06 Jan '12

I am attempting to get Janino 2.6.1 working with Logback 0.9.28 (or later, but that's the version we are using in Eclipse Virgo right now) - see [0] for background. Unfortunately Janino ends up using the thread context class loader as its "parent" class loader and fails with a runtime exception. I have discussed this ([1]) on the Janino mailing list and it seems that it is necessary to set the parent class loader in Janino. I am somewhat at the mercy of Logback here. The surprising thing is that the code in Logback 0.9.24 looked pretty usable in this respect. JaninoEventEvaluator.start had the following sequence: ClassLoader cl = context.getClass().getClassLoader(); ee = new ExpressionEvaluator(getDecoratedExpression(), EXPRESSION_TYPE, getParameterNames(), getParameterTypes(), THROWN_EXCEPTIONS, cl); thus setting the parent class loader to a value which I could ensure would be capable of loading the necessary types. In 0.9.28 this code has been replaced by: scriptEvaluator = new ScriptEvaluator(getDecoratedExpression(), EXPRESSION_TYPE, getParameterNames(), getParameterTypes(), THROWN_EXCEPTIONS); which causes the current TCCL to be used as the parent class loader and ultimately results in Janino failing. I can't control the TCCL that happens to be in use when start is called as that is driven out of a logging call which can come from an arbitrary thread. I found the relevant commit ([2]), but I can't tell from that why this specific change was made. If it is absolutely necessary to use ScriptEvaluator rather than ExpressionEvaluator, the following code sequence (based on a suggestion from Arno Unkrig) would reproduce the parent class loader behaviour of 0.9.24: ClassLoader cl = context.getClass().getClassLoader(); scriptEvaluator = new ScriptEvaluator(); scriptEvaluator.setParentClassLoader(cl); scriptEvaluator.setReturnType(EXPRESSION_TYPE); scriptEvaluator.setParameterNames(getParameterNames()); scriptEvaluator.setParameterTypes(getParameterTypes()); scriptEvaluator.setThrownExceptions(THROWN_EXCEPTIONS); scriptEvaluator.cook(getDecoratedExpression()); The alternative of using ExpressionEvaluator is much neater and seems to be close to the ScriptEvaluator variant since ExpressionEvaluator extends ScriptEvaluator. Any suggestions gratefully received. Regards, Glyn [0] https://bugs.eclipse.org/bugs/show_bug.cgi?id=333920 [1] mailing list archive link currently broken - if I can get a link to it, I'll post this later [2] https://github.com/ceki/logback/commit/06a5b692f14560636bd92d7bd7cf1f85830f…

2 2

Issues configuring GMail appender
by Greg Silin 04 Jan '12

04 Jan '12

Hi, I'm trying to setup logback for a fairly simple appender configuration.. DEBUG, INFO, WARN: System.out Console Appender, File Appender ERROR: System.err Console Appender, File Appender, SMTPAppender configured for GMail. Following instructions to setup GMail appender, I have the following: Appender config:  <appender name="MailAppender" class="ch.qos.logback.classic.net.SMTPAppender"> <smtpHost>smtp.gmail.com</smtpHost> <smtpPort>465</smtpPort> <ssl>true</ssl> <username>MyUsername</username> <password>MyPassword</password> <to>greg(a)klout.com</to>  <from>noreply(a)gmail.com</from> <subject>LOGGED ERROR: %logger{20} - %m</subject> <layout class="ch.qos.logback.classic.PatternLayout"> <pattern>%date %-5level %logger{35} - %message%n%xEx{35}</pattern> </layout> </appender> Appender inclusion: <root level="ERROR"> <appender-ref ref="StdErrAppender" /> <appender-ref ref="FileAppender" /> <appender-ref ref="MailAppender" /> </root> All appropriate maven dependencies: logger jars <slf4j.version>1.6.4</slf4j.version> <dependency> <groupId>org.slf4j</groupId> <artifactId>slf4j-api</artifactId> <version>${slf4j.version}</version> </dependency> <dependency> <groupId>ch.qos.logback</groupId> <artifactId>logback-access</artifactId> <version>1.0.0</version> </dependency> <dependency> <groupId>ch.qos.logback</groupId> <artifactId>logback-classic</artifactId> <version>1.0.0</version> </dependency> <dependency> <groupId>ch.qos.logback</groupId> <artifactId>logback-core</artifactId> <version>1.0.0</version> </dependency> java mail <dependency> <groupId>javax.mail</groupId> <artifactId>mail</artifactId> <version>1.4.4</version> </dependency> All the other appenders work just fine as configured. I see that the SMTPAppender is being loaded w/o error (at least nothing I see reported) .. yet on ERROR nothing gets sent. Username & password are all correct. I've confirmed with our operations that we don't have any special spam filters set up. Thoughts on what could be wrong / what I need to check? Thanks -greg

2 1