Several parts of the zip package handle the UTF-8 encoding 
of entry names incorrectly. They assume either an ancient form of UTF-8 
which didn't have the 4-byte form for supplementary characters, or rely 
on the JVM's modified UTF-8, which has the same limitation. As a 
consequence, file names using supplementary characters can be used, but 
cannot be exchanged with standards-compliant zip implementations.

The following parts of the implementation are incorrect:
src/share/classes/java/util/zip/ZipInputStream.java
     getUTF8String
src/share/classes/java/util/zip/ZipOutputStream.java
     getUTF8Length
     getUTF8Bytes
src/share/native/java/util/zip/ZipEntry.c
     Java_java_util_zip_ZipEntry_initFields
src/share/native/java/util/zip/ZipFile.c
     Java_java_util_zip_ZipFile_getEntry
     Java_java_util_jar_JarFile_getMetaInfEntryNames
======================================================================
Posted Date : 2005-08-21 05:24:28.0

N/A

Yes, we need to fix these to improve support for supplementary characters.

  xxxxx@xxxxx   2004-04-12
Changing the current encoding to support 4-byte supplementary characters could result in creating JAR files that are incompatible, i.e. cannot be read, by previous Java releases.  This incompatibility is not acceptable.  In fixing 4244499 though, there is a reasonable chance that support can be provided for the current implementation as well as standard UTF-8.
Posted Date : 2008-04-09 21:20:03.0

We go with the standard UTF-8 chaset. We can NOT keep the forward-compatibility for this case. If someone really needs to generate the "old-style" jar/zip file with the latest version of JDK/JRE, we might consider to add the "modified" utf-8 into our charset respository.
Posted Date : 2009-04-17 04:19:34.0

Yes fix it!!
Pls!
I have been reading and testing for about a day for this 
issue! and it's here since v1.1

So.. I've searched this bug but I've found anything...
Do you have new information about this bug ?

This is probably a dup of or at least related to 4244499.