The version of Apache log4j used by SoundHelix.

[[ 🗃 ^aEp6o apache-log4j ]] :: [📥 Inbox] [📤 Outbox] [🐤 Followers] [🤝 Collaborators] [🛠 Commits]

Clone

HTTPS: git clone https://vervis.peers.community/repos/aEp6o

SSH: git clone USERNAME@vervis.peers.community:aEp6o

Branches

Tags

v_1_0 ::

TROUBLESHOOT.html

<html>
<head>
<title>Troubleshooting log4j</title>
</head> 
<body bgcolor=#FFFFFF>
 
<h1 align=center>Log4j troubleshooting</h1>

<h2 align=center>Ceki G&uuml;lc&uuml;
<br>November 2000</h2>

<hr>

<p>Here is a list of commonly encoutered problems when using log4j:

<ul> 

<li><p><a href=#noAppenders>log4j tells me to initialize
properly.</a>

<li><p><a href=#duplicates>Duplicates in log4j output.</a>

<li><p><a href=#space>Options are not parsed correctly.</a>

<li><p><a href=#jit>Location information is printed as a "?" character.</a>

<li><p><a href=#cce>ClassCastException when instantiating a Category subclasses.</a>

</ul>

<hr>

<p><a name=noAppenders><h4>log4j tells me to initialize properly.</h4>

Logging output is written to a target by using an appender. If no
appenders are attached to a category nor to any of its ancestors, you
will get the following message when trying to log:

<pre>
log4j: No appenders could be found for category (some.category.name).
log4j: Please initialize the log4j system properly.
</pre>

<em>Log4j does not have a default logging target.</em> It is the
user's responsability to ensure that all categories can inherit an
appender.  This can be easily achieved by attaching an appender to the
root category.


<p><a name=duplicates><h4>Duplicates in log4j output.</h4>

<p>The reason for observing duplicates in log4j output is either due
to having added the same appender multiple times to the same category
(typically root) or having added the same appender to different
categories not knowing that appenders are inherited cumulatively.

<p>log4j does not eliminate appender duplicates. In other words, if
you add the same appender to a categoty <i>n</i> times, that appender
will be invoked <i>n</i> times to append to its target. 

<p>A slightly different cause is adding different appenders all
sharing the same underlying output target to some category. In the
most common occurance of this phenomenon, the
BasicConfigurator.configure() method is invoked multiple times. Each
time it is invoked, this method adds an appender with a
<code>System.out</code> target to the root category. See <a
href="http://sourceforge.net/support/?func=detailsupport&support_id=107779&group_id=2666">[
Support #107779]</a> for an example.

<p>One other common mistake is to forget that appenders are inherited
cumulatively from the hierarchy. For example, if you add an appender,
say <code>A</code>, to the root category, all other categories will
inherit <code>A</code> as an appender. Thus, if you add <code>A</code>
to a categoy, say <code>C</code>, then an enabled statement of
category <code>C</code>, will print to <code>A</code> twice, once
because <code>A</code> is in root and once because it is in
<code>C</code>. See <a
href="http://sourceforge.net/bugs/?func=detailbug&bug_id=121892&group_id=2666">[
Bug #121892 ]</a> for an example.

<p><a name=space><h4>Options are not parsed correctly.</h4>

The PropertyConfigurator relies on <code>java.util.Properties</code>
class to read in the configuration file. This class preserves spaces
in options. For example,

<pre>
fruit=orange  
</pre>
is returned as an option having the key <code>"fruit"</code> and the
value <code>"orange "</code>.

<p>The spaces in the value, i.e. <code>"orange "</code>, are due to
invisible spaces at the end of the example shown above. Thus, some of
the options might not be interpreted correctly due to trailing
spaces. See <a
href="http://sourceforge.net/bugs/?func=detailbug&bug_id=121903&group_id=2666">[Bug
#121903]</a> for an example of this problem.

<p>In log4j version 0.9.0, all spaces are removed from <em>both</em>
ends of option values. In version 0.9.1 log4j reverted to the old
behaviour where option values are not all automatically trimmed. 


<p><a name=jit><h4>Location information is printed as a "?" character.</h4>

Location information is extracted automatically by the PatternLayout
conversion patterns %C, %F, %M and %L. However, some just-in-time
(JIT) compilers make it impossible to extract location information. It
is also possible that the complier that generated the byte code may
have ommitted the LineNumber table as is done by -O option of javac
and jikes.  

<p>You can remedy this problem by disabling the JIT compiler and by
compiling the code without the -O option.

<p>See <a href="http://sourceforge.net/support/?func=detailsupport&group_id=2666&support_id=108314">[Support
#108314]</a> for an example of this problem.

<p><a name=cce><h4><code>ClassCastException</code> when instantiating
a <code>Category</code> subclasses.</a></h4>

<p>This exception is thrown because log4j does not support
homonyms. For example, the following will systematically throw a
<code>ClassCastException</code>

<pre>
  Category c1 = Category.getInstance("bad");
  MyCategory c2 = (MyCategory) MyCategory.getInstance("bad");
</pre>

where <code>MyCategory</code> is a sub-class of
<code>Category</code>. The problem occurs because the second
<code>getInstance</code> invocation will retrieve the category created
in the fist invocation. This instance is a <code>Category</code>
object and cannot be cast to <code>MyCategory</code>.

<p>By default, the <code>PropertyConfigurator</code> will create and
configure <code>org.log4j.Category</code> objects. Thus, if you try to
instantiate a category sub-class for an already existing category, and
try to cast it to the sub-class type, you will systematically get a
<code>ClassCastException</code>.

To address this problem, the <code>PropertyConfigurator</code> admits the 

</body>
</html>
[See repo JSON]