Logback

Contents
1.Logging

2.Introduction to slf4j logging API

3.Logback vs. Log4j

4.About Logback

5.Logback Architecture

6.Components of Logback

7.Configuration of Logback

8.Basic logback.xml

9.Convenience with logback

10.Conventions with logback

11.Links

 Logging as we all know is a way of recording events at an appropriate output
store.
 Based on convenience we can get the output at the console, file, database or
email.
 The process is capable of stating the flow of execution and can even be helpful
to locate
the error points in an application.

Introduction to slf4j logging API.
• SLF4J is only a facade, meaning that it does not provide a complete logging solution.
• With any other logging solution working at the back end, slf4j acts a consistent logging
system and keeps the user unknown about which logging mechanism is being used.

• slf4j.zip ships with bindings for all logging mechanisms. User can switch to any logging
just by changing to the corresponding jar file.

Logback vs Log4j
Log4j Logback
Can be directly implemented in Logback natively speaks slf4j i.e, it exposes
applications. it’s API via slf4j.

Log4j can be configured using xml file as Logback can be configured in a xml file and
well as properties file. groovy. There are online application to
convert an existing log4j.properties to a
logback.xml file.

Once log4j is implemented at the code level Logback project can be shifted to any other
Its becomes difficult to move to any other logging mechanism easily only by removing
logging mechanism. logback.jar and placing the other required
slf4j binding.

About Logback
• Logback is intended as a successor to the popular log4j project.
• It is no revolution but evolution, as the author of logback Sébastien Pennec says.
• It edges log4j as log4j is merely developer’s logging system to trace the execution
of a program whereas logback is even capable of generating server side logs when
a use accesses a webpage on the server.
• Logback exposes it’s API via slf4j and can be easily replaced by any other logging
mechanism depending upon the use case scenario.
• In some situations it has even been upto 10 times efficient than log4j.

Architecture of logback
Logback in its current release is divided
into three modules.
• Logback-core : It lays the ground work
for the other two modules. Acts as the
super class to some of the classes in the
other modules.
• Logback-classic : Logback classic
extends the core module and internally
implements slf4j so that the user can
readily switch back and fourth between
various other logging mechanisms.
• Logback -access : Logback –access
integrates with Servlet containers to
provide HTTP-access log functionality.
• Slf4j acts as the interface to various
other logging mechanisms.

Components of Logback.
The main components of logback are as follows.
• Appender
• Encoder
• Layout
• Filter
• Logging Levels

• Logback appender is a part of logback-core module.
• Logback delegates the task of writing a logging event to components called appenders
• Different appenders override doAppend(Event e) method and write the events to the
specified appender.

Layout: Layout are capable to accept events and write those events as a strings.
 Until Logback 0.9.19 appenders directly relied on layouts to write the event at
the appropriate output destination.
 Layouts had no control on when the event has to be written out so could not
aggregate events into batches.
 Accepts the format in which the output has to be written out.
Encoder: They accept events and transform these events directly into byte array
and
write into the output stream.
 Binds a layout into it that specifies the format of the output.
 Unlike layout has total control over when the data has to be written out.

Filter: These are the logback-classic components that are capable of filtering the
incoming events based on the specified rule and can deny or accept the
event for being printed.
ex. <level>debug</level>
<onMatch>Accpet</onMatch>
<onMismatch>Deny</onMismatch>

Logback offers six levels of
Logging. They are as mentioned Effectiveness

below:-
 Trace Requested Trace Debug Info Warn Error Off
level
 Debug
Trace yes no no no no no
 Info
 Warn
 Error Debug yes yes no no no no
 Off
The table shows each level and
Info yes yes yes no no no
their effectiveness.

Warn yes yes yes yes no no

Error yes yes yes yes yes no

 Logback can be configured into a .xml file or .groovy file.
 During the initialization of logback in an application it tries to load the
logback.groovy file.
 In the absence of logback.groovy file, logback looks for logback-test.xml file and
if not available it search for logback.xml file to configure itself.
 If the above mentioned file is not present in the application,logback configures
itself using BasicConfigurator which will cause all the logging statements to be
directed to the console.

Lets see how to configure all these Logback components in an xml file.

Given below are some sample logback.xml files highlighting the use of
appender,encoder,pattern,Logging levels and logger.

 Logback-classic always checks for any modification in the configuration and is
capable of applying the changes if found dynamically.
ex - <configuration scan="true" scanPeriod="10 seconds" debug="true">
 Logback allows parameterized logging statement.
ex – This approach eliminates the overhead of creating extra string in the below mentioned
statement.
String name = “Anubhav”;
if(debugEnabled)
{logger.debug(“the name to be printed is”+name);}
Instead it can be
String name = “Anubhav”;
Logger.debug(“The name to printed is {}”,name);

 A developer can even trace the internal status of the logback processing by
calling upon the LoggerContext and printing its status as follows.
LoggerContext lc = (LoggerContext) LoggerFactory.getILoggerFactory();
StatusPrinter.print(lc);

 Directory structures can be mentioned as properties tags in the xml file to avoid
retyping
everytime.
ex -<properties name =”Logback_Dir” value=”d:/logFiles”>
Can be referenced as
<file>${Logback_Dir}/logFile.log</file>
 Resource files can also be configured to be loaded and the keys can be
referenced from
logback.xml file.
ex- <property file=”src/MessageResource.properties.”>.
The key can be referenced as
<mail_to>${MailTo}</mail_to>

 Let the declaration of logger be always the first statement inside a class.
 Use trace level only when tracing the whole application is necessary as it
activates
all levels.
 While entering a function the logger statement should look like
logger.debug(“>>Entering the function”);
 While exiting the function the logger statement should look like
logger.debug(“<<Exiting the function”);
 The instantiation of an object should be captured as
logger.debug(“** new ClassConstructor()”);
 When inside a method the log statement should always contain the method
name
logger.debug(“myFunction()>the parameter is {}”,parameter);


Use appropriate tool to shift log4j configured project to slf4j.
ex- slt4j migrator. it will replace appropriate import lines and logger declarations.

Place slf4j.jar,logback-core.jar and logback-classic.jar in the classpath.

Use the following tool to convert log4j.properties to logback.xml and deploy the
xml file.
Link - http://logback.qos.ch/translator/
Shift Complete

 http://logback.qos.ch/
 http://logback.qos.ch/documentation.html
 http://logback.qos.ch/manual/index.html
 http://logback.qos.ch/manual/appenders.html
 http://logback.qos.ch/manual/encoders.html
 http://logback.qos.ch/manual/layouts.html
 http://logback.qos.ch/manual/filters.html
 https://wiki.base22.com/display/btg/Java+Logging+Standards+and+Guideli
nes

Logback

Recommended

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Logback (20)

Logback