I ran into a situation at work the other day where I wanted to disect some rather large Log4J log files. Each of the files contained a bunch of text log entries of the following format:

13:20:32,243 INFO [localhost-startStop-1][DialectDetector:71] Determine dialect

What I wanted was a tool to help visualize what was going on in these files. After a quick search, I stumbled upon two free tools I thought may do the trick:

Chainsaw

Chainsaw is “a companion application to Log4j written by members of the Log4j development community”. In Chainsaw, I had to create a new Receiver in order to process the .log file. The receiver designates a “logFormat” property in which you can specify the pattern for each log message. After some trial and error, these settings seemed to work:

 fileURL: file:\\\C:\<My Log Location>\<My log>.log
 filterExpression:
 logFormat: TIMESTAMP LEVEL [THREAD][CLASS:LINE] MESSAGE
 name: Custom Receiver
 tailing: false
 timestampFormat: HH:mm:ss,SSS

Otros Log Viewer

Otros Log Viewer is an open source piece of software whose intention is to assist in the analysis of application logs and traces. In order to establish a connection between Otros Log Viewer and the log file I was attempting to analyze, I needed to supply a definition for each log file entry. To do this, I created a .pattern file in the plugins > logimporters directory which contained the text:

 type=log4j
 pattern=TIMESTAMP LEVEL [THREAD][CLASS:LINE] MESSAGE
 dateFormat=HH:mm:ss,SSS
 name=Lo4j-CustomReceiver
 charset=windows-1252