I have a working Java program that uses the Tanuki wrapper. The problem I have is that the wrapper log file is not UTF-8 encoded, but appears to be ASCII. The wrapper configuration file begins with:
#encoding=UTF-8
#include ..\..\Tomcat\conf\wrapper-license.conf
wrapper.java.command.loglevel=INFO
wrapper.lang.encoding=UTF-8
wrapper.debug=true
The wrapper starts and it starts the JVM successfully. But when I edit the wrapper logfile, Japanese characters (for example) are present as question mark characters, i.e., ASCII character 0x3f. I double-checked by loading the log file in a hex editor.
The Tanuki Wrapper log file is put into a directory that contains Japanese characters -- for testing purposes. The log file is successfully created in that folder, so the wrapper is clearly able to read and process the UTF-8 characters. But when it logs the folder name in which it will create its logs, the folder name is logged as all ASCII 0x3f characters ('?').
How can I get the Tanuki Wrapper to encode its log file in UTF-8?
I have confirmation from Tanuki that the current wrapper software will always write its logs using the current system encoding. There is currently (as of 3.5.17) no way to configure the wrapper to write its logs in any different encoding, such as UTF-8.
Again, you can configure the encoding in which the wrapper will read the configuration file, but not the encoding with which it writes to its log file.