- logback-user - mailman3.qos.ch

Are special charcters allowed in logger configurations?
by Noman Amir Jamil 04 Jan '12

04 Jan '12

Hi, I would like to know if the following issue is resolved in logback, it was previously encountered in log4j: "there are known issues with using this naming convention (with square brackets) in log4j XML based configuration files" I am using a similar configuration with square brackets: logger name="org.apache.catalina.core.ContainerBase.[Catalina].[localhost]" level="INFO" additivity="false"> But it appears that the portion after ContainerBase is being ignored or not processed. I would really appreciate any help on this. Thanks Noman A.

2 4

SMTPAppender OSGi-Problem within Glassfish v3.1.1 (Apache Felix)
by cinhtau 04 Jan '12

04 Jan '12

Hi @all, I try to use SMTPAppender for logging errors to respective users via E-Mail in OSGi. My configuration is: OS: Ubuntu Linux 10.04.03 LTS Application Environment: Glassfish 3.1.1 using Apache Felix Logback: tried logback classic Version 0.29 and current Version 1.0 leading to same results Glassfish has following bundles g! lb mail START LEVEL 10 ID|State |Level|Name 227|Resolved | 1|JavaMail API (1.4.4) I write following BundleActivator (I switch the real addresses with stubs), sending an email via SMTPAppender and Java Mail from scratch. package de.sgbs.log; import java.util.logging.Level; import javax.mail.Message; import javax.mail.MessagingException; import javax.mail.Session; import javax.mail.Transport; import javax.mail.internet.InternetAddress; import javax.mail.internet.MimeMessage; import org.osgi.framework.BundleActivator; import org.osgi.framework.BundleContext; import org.slf4j.Logger; import org.slf4j.LoggerFactory; import org.slf4j.Marker; import org.slf4j.MarkerFactory; import ch.qos.logback.classic.LoggerContext; import ch.qos.logback.classic.joran.JoranConfigurator; import ch.qos.logback.core.joran.spi.JoranException; public class Activator implements BundleActivator { private final static Logger logger = LoggerFactory.getLogger(Activator.class); @Override public void start(BundleContext context) throws Exception { try { LoggerContext lc = (LoggerContext) LoggerFactory.getILoggerFactory(); JoranConfigurator configurator = new JoranConfigurator(); configurator.setContext(lc); lc.reset(); lc.putProperty("application-name", Activator.class.getSimpleName()); configurator.doConfigure("logback.xml"); Marker notifyAdmin = MarkerFactory.getMarker("NOTIFY_ADMIN"); logger.error(notifyAdmin, "This is a serious an error requiring the admin's attention", new Exception("Just testing")); } catch (JoranException ex) { java.util.logging.Logger.getLogger(Activator.class.getName()).log(Level.SEVERE, null, ex); } sendEMail(); } private void sendEMail() throws MessagingException { java.util.Properties props = new java.util.Properties(); props.put("mail.smtp.host", "smtp"); props.put("mail.smtp.port", "25"); Session session = Session.getDefaultInstance(props, null); // Construct the message Message msg = new MimeMessage(session); msg.setFrom(new InternetAddress("me(a)gmail.com")); msg.setRecipient(Message.RecipientType.TO, new InternetAddress("foo(a)bar.com")); msg.setSubject("Test"); msg.setText("Hello user, you got an error:"); // Send the message Transport.send(msg); } @Override public void stop(BundleContext context) { logger.info("Goodbye Community!"); } } My logback.xml NOTIFY_ADMIN TRANSACTION_FAILURE smtp foo(a)bar.com me(a)gmail.com TESTING: %logger{20} - %m %date %-5level %logger{35} - %message%n Sending from scratch works. SMTPAppender raises an error. This error is given on the osgi console: 10:14:46,850 |-INFO in ch.qos.logback.core.joran.action.AppenderRefAction - Attaching appender named [EMAIL] to Logger[ROOT] 10:14:47,048 |-ERROR in ch.qos.logback.classic.net.SMTPAppender[EMAIL] - Error occured while sending e-mail notification. javax.mail.MessagingException: IOException while sending message; nested exception is: javax.activation.UnsupportedDataTypeException: no object DCH for MIME type multipart/mixed; boundary="----=_Part_5_17758761.1325668486851" at javax.mail.MessagingException: IOException while sending message at at com.sun.mail.smtp.SMTPTransport.sendMessage(SMTPTransport.java:1141) at at javax.mail.Transport.send0(Transport.java:195) at at javax.mail.Transport.send(Transport.java:124) at at ch.qos.logback.core.net.SMTPAppenderBase.sendBuffer(SMTPAppenderBase.java:343) at at ch.qos.logback.core.net.SMTPAppenderBase.append(SMTPAppenderBase.java:179) at at ch.qos.logback.core.AppenderBase.doAppend(AppenderBase.java:85) at at ch.qos.logback.core.spi.AppenderAttachableImpl.appendLoopOnAppenders(AppenderAttachableImpl.java:64) at at ch.qos.logback.classic.Logger.appendLoopOnAppenders(Logger.java:285) at at ch.qos.logback.classic.Logger.callAppenders(Logger.java:272) at at ch.qos.logback.classic.Logger.buildLoggingEventAndAppend(Logger.java:473) at at ch.qos.logback.classic.Logger.filterAndLog_0_Or3Plus(Logger.java:427) at at ch.qos.logback.classic.Logger.error(Logger.java:610) at at de.sgbs.log.Activator.start(Activator.java:37) at at org.apache.felix.framework.util.SecureAction.startActivator(SecureAction.java:629) ... Caused by: javax.activation.UnsupportedDataTypeException: no object DCH for MIME type multipart/mixed; boundary="----=_Part_5_17758761.1325668486851" at at javax.activation.ObjectDataContentHandler.writeTo(DataHandler.java:877) at at javax.activation.DataHandler.writeTo(DataHandler.java:302) at at javax.mail.internet.MimeBodyPart.writeTo(MimeBodyPart.java:1476) at at javax.mail.internet.MimeMessage.writeTo(MimeMessage.java:1772) at at com.sun.mail.smtp.SMTPTransport.sendMessage(SMTPTransport.java:1099) at ... 47 common frames omitted 10:14:47,049 |-INFO in ch.qos.logback.classic.net.SMTPAppender[EMAIL] - SMTPAppender [EMAIL] is tracking [1] buffers First I thought MIME type multipart/mixed is the cause. I checked my SMTP Server Settings, sending email is ok. I tried googlemail. It's working too. I found another https://bugs.eclipse.org/bugs/show_bug.cgi?id=322398 resource with the same error message. Could it be an classloader problem? Could someone help me to pinpoint the problem? Thanks in advance. -- View this message in context: http://old.nabble.com/SMTPAppender-OSGi-Problem-within-Glassfish-v3.1.1-%28… Sent from the Logback User mailing list archive at Nabble.com.

2 2

logback in WebLogic 11g/12c
by Thai Dang Vu 02 Jan '12

02 Jan '12

Hi All, My application is in an exploded EAR directory which has a war and a jar module. Its directory structure can be seen in the attached fig1.png (where log4j and commons-logging are used) and fig4.png (where logback is used). The application.xml is in fig2.png. The log4j.properties and logback.xml are in the jar module so that other modules (war, ejb) can use it. The servlet (fig3.png) writes to the log. The application works fine with log4j, but nothing is printed out in case of logback. The application is in again.zip. The exploded EAR directory can be generated with running "mvn clean package" in the root directory. Does anybody know where I'm wrong with logback? Regards.

1 0

Time and Size Based Rolling Not Working
by Noman Amir Jamil 31 Dec '11

31 Dec '11

Hi, I have recently switched to logback for tomcat6 logging. I was able to successfully set it up and log files are being created and logged into. Following is my logback.xml: ================= <?xml version="1.0" encoding="UTF-8"?> <configuration> <appender name="CONSOLE" class="org.apache.juli.logging.ch.qos.logback.core.ConsoleAppender"> <encoder> <pattern>%d{HH:mm:ss.SSS} %-5level {%thread} [%logger{20}] : %msg%n</pattern> </encoder> </appender> <appender name="FILE-CATALINA" class="org.apache.juli.logging.ch.qos.logback.core.rolling.RollingFileAppender"> <file>${catalina.base}/logs/catalina.log</file> <append>true</append> <encoder> <charset>utf-8</charset> <pattern>%d{HH:mm:ss.SSS} %-5level {%thread} [%logger{40}] : %msg%n</pattern> </encoder> <rollingPolicy class="org.apache.juli.logging.ch.qos.logback.core.rolling.TimeBasedRollingPolicy"> <fileNamePattern>${catalina.base}/logs/catalina-%d{yyyyMMdd}-%i.log.zip</fileNamePattern> <maxHistory>60</maxHistory> <timeBasedFileNamingAndTriggeringPolicy class="org.apache.juli.logging.ch.qos.logback.core.rolling.SizeAndTimeBasedFNATP"> <maxFileSize>10KB</maxFileSize> </timeBasedFileNamingAndTriggeringPolicy> </rollingPolicy> </appender> <appender name="FILE-LOCALHOST" class="org.apache.juli.logging.ch.qos.logback.core.rolling.RollingFileAppender"> <file>${catalina.base}/logs/localhost.log</file> <append>true</append> <encoder> <charset>utf-8</charset> <pattern>%d{HH:mm:ss.SSS} %logger{0} {%thread} %level : %msg%n</pattern> </encoder> <rollingPolicy class="org.apache.juli.logging.ch.qos.logback.core.rolling.TimeBasedRollingPolicy"> <fileNamePattern>${catalina.base}/logs/localhost-%d{yyyyMMdd}-%i.log.zip</fileNamePattern> <maxHistory>60</maxHistory> <timeBasedFileNamingAndTriggeringPolicy class="org.apache.juli.logging.ch.qos.logback.core.rolling.SizeAndTimeBasedFNATP"> <maxFileSize>10KB</maxFileSize> </timeBasedFileNamingAndTriggeringPolicy> </rollingPolicy> </appender> <logger name="org.apache.catalina" level="INFO" additivity="false"> <appender-ref ref="FILE-CATALINA" /> </logger> <logger name="org.apache.catalina.core.ContainerBase.[Catalina].[localhost]" level="INFO" additivity="false"> <appender-ref ref="FILE-LOCALHOST" /> </logger> <root level="INFO"> <appender-ref ref="CONSOLE" /> </root> </configuration> ================= The logback jars being used are of version logback-0.9.30 as higher version doesn't work with tomcat6. The issue is that log files for localhost appender are being rolled over, but the catalina logs are not rolled over even after reaching the specified limit, unless I restart tomcat. The tomcat6 is running on Windows Server 2008. Can anybody help me with this? Thanks Noman A.

1 0

Re: [logback-user] Prudent [sic!] mode logging to NFS-mounted file
by ThomasSchuessler2＠Eaton.com 29 Dec '11

29 Dec '11

Thank you very much for the fast answer. I feel so stupid. -----Original Message----- From: logback-user-bounces(a)qos.ch [mailto:logback-user-bounces@qos.ch] On Behalf Of ceki Sent: Mittwoch, 21. Dezember 2011 17:49 To: logback users list Subject: Re: [logback-user] Prudent mode logging to NFS-mounted file Check spelling of prudent (not prudend). On 21.12.2011 16:30, ThomasSchuessler2(a)Eaton.com wrote: > We seem to be losing quite a few log records in a clustered web > application writing to an NFS-mounted file with the configuration > shown below. > > We are running logback-1.0.0. Any suggestions? > > <appendername=/"file"/ > > class=/"ch.qos.logback.core.rolling.RollingFileAppender"/> > <append>true</append> > <prudend>true</prudend> > <encoder> > <pattern>%d ${HOSTNAME} [%t] %-5p %c - {%marker} %m%n</pattern> > </encoder> -- Ceki http://twitter.com/#!/ceki _______________________________________________ Logback-user mailing list Logback-user(a)qos.ch http://mailman.qos.ch/mailman/listinfo/logback-user --------------------------------------------------------------------------

4 4

syslogAppender, suffixPattern, "valid conversion word"
by Ted Timmons 29 Dec '11

29 Dec '11

Hi. I have a file-based logback in place. I'm adding syslogging to it. Here's a sample configuration: <appender name="SYSLOG-MAIN" class="ch.qos.logback.classic.net.SyslogAppender"> <syslogHost>remotehost</syslogHost> <facility>local6</facility>    </appender> It works with the suffixPattern lines commented out. If I use the first line ("hello world"), I get the following: 01:46:35,272 |-INFO in ch.qos.logback.core.joran.action.AppenderAction - About to instantiate appender of type [ch.qos.logback.classic.net.SyslogAppender] 01:46:35,276 |-INFO in ch.qos.logback.core.joran.action.AppenderAction - Naming appender as [SYSLOG-MAIN] 01:46:35,293 |-ERROR in ch.qos.logback.core.pattern.parser.Compiler@2db7ae22 - There is no conversion class registered for conversion word [nopexsyslog] 01:46:35,293 |-ERROR in ch.qos.logback.core.pattern.parser.Compiler@2db7ae22 - [nopexsyslog] is not a valid conversion word Okay, I think I understand "nopex". But why isn't plain text acceptable? I get similar errors for the next two lines, which is what I'd actually like the logging to appear as. 01:20:12,978 |-ERROR in ch.qos.logback.core.pattern.parser.Compiler@31f9338f - There is no conversion class registered for conversion word [nopexatime] 01:20:12,978 |-ERROR in ch.qos.logback.core.pattern.parser.Compiler@31f9338f - [nopexatime] is not a valid conversion word I'm running logback-0.9.28. Thanks, -ted

1 0

Setting custom Profiler log level.
by César Álvarez Núñez 27 Dec '11

27 Dec '11

Hi all, Our application makes an intensive use of Profiler ( http://www.slf4j.org/extensions.html#profiler) It has been very useful during the development stage an we would like to keep it at production stage but there is a problem since its "log level" is hard-coded to "debug". Currently the profiler use the same application class logger so if log level is set to "info" in our production environment we lose the "profiler" log :-( Alternatives: - Use different loggers for application and profiler >-->> It requires to refactoring all the code and remember that any new Profiler that will be used need to set a custom logger instead of the default class logger. - Open a request to modify Profiler in order to set a custom log level. - Keep minimum log level to "debug" and make use of a Filter to accept any log event with marker=PROFILER, otherwise only accept if log level >= INFO >-->> It only will work with LogBack but Profiler belongs to SLF4J so it should work with any bridge. Any other way? BR, /César.

3 4

Best strategy for centralized logs in a multitenant environment
by Xavi Beumala 26 Dec '11

26 Dec '11

Hi all, I'm working on a multitenant application where I'd like to have different log files for each customer. I've been looking at the SiftingAppender and it seems to perfectly fit my needs. The application will run on a sessionless cluster, there isn't any kind of shard, each customer request can run on any node. Ideally I'd like to have a centralized log file for each customer and have the ability to hot change its configuraiton. What configuration would you better recommend? - DB based appender? - File based appender running prudent mode? - I've seen the scan="true" option to automatically reload the logging configuration. Ideally I'm looking for an option where automatically a configuration file could be hot plugged per tenant / customer. Is there any option so that scan=true checks a directory hierarchy? Many thanks for your help and recomendations. Much appreciated Xavi

1 0

DBAppender - can I use my own tables?
by Nadia Kunkov 23 Dec '11

23 Dec '11

Hello everyone - I wanted to use the logback DBAppender but in the documentation it says that it logs to a set of 3 specific tables. Is it possible to change the tables and use some other ones? Nadia

4 6

logback console appender slow on windows xp
by Peter Kullmann 22 Dec '11

22 Dec '11

Hi, we had a problem with the performance of our application and it boiled down to very bad performance of the console appender when there is no console (in particular on windows xp). You can see the whole context of the issue here: http://www.eclipse.org/forums/index.php/m/768740/#msg_768740 Here's the conclusion and the code for my experiments: package testlogback; import java.io.FileWriter; import java.io.IOException; import org.slf4j.Logger; import org.slf4j.LoggerFactory; public class Main { public static void main(String[] args) throws IOException { Logger logger = LoggerFactory .getLogger("chapters.introduction.HelloWorld1"); long start = System.currentTimeMillis(); for (int i = 0; i < 10000; i++) { logger.debug("Hello world."); } long end = System.currentTimeMillis(); FileWriter fileWriter = new FileWriter("logbacktest.out"); fileWriter.write("Time: " + (end - start) + "\n"); fileWriter.close(); System.out.println("Time: " + (end - start)); } } On my windows box it prints out "Time: 140" when started as "java -jar testlogback.jar". If I start "javaw -jar testlogback.jar" it uses 100% CPU for about 20 seconds (for effectively doing nothing). This is the case for Sun ire 6 and 7 on windows xp. On windows 7 the situation is better: javaw uses only about 7 seconds (which is still too much, I guess). Best regards, Peter

2 9