Actually, I'm embarrassed to say this, but I just realized that I didn't have the CXF jars in my client's classpath, just the JAX jars, so it worked okay with "proper" messages but didn't know what to do with faults. Putting the CXF jars (I didn't know exactly which were needed, so I took the whole distribution) in place worked.
-----Original Message----- From: Daniel Kulp [mailto:[EMAIL PROTECTED] Sent: Thursday, March 20, 2008 2:18 PM To: [email protected] Cc: Michael Berkowitz Subject: Re: When client gets fault response, it throws "java.lang.NoClassDefFoundError: Could not initialize class com.sun.xml.internal.ws.fault.SOAPFaultBuilder" On Wednesday 19 March 2008, Michael Berkowitz wrote: > Thanks, Daniel. I am, in fact, using build 1.6.0_04-b12. Thanks for > the advice, I'll look into it. What's the ETA on 2.1? I started a discussion about that: http://www.nabble.com/Options-for-CXF-2.1.....-to16158740.html Users inputs would be welcome as well. :-) Dan > -----Original Message----- > From: Daniel Kulp [mailto:[EMAIL PROTECTED] > Sent: Wednesday, March 19, 2008 4:05 PM > To: [email protected] > Cc: Michael Berkowitz > Subject: Re: When client gets fault response, it throws > "java.lang.NoClassDefFoundError: Could not initialize class > com.sun.xml.internal.ws.fault.SOAPFaultBuilder" > > > > Looking at the error, I guess it's Java 6? > > That's probably the "issue". My gut feeling says it would work ok if > you used a Java5 vm. Getting 2.0.x to completely work on Java 6 (up > to > > update 3, but not update 4) can be done, but you probably need to copy > a > > couple jars from the lib dir to the jre/lib/endorsed dir. > Specifically, the saaj* jars and the jaxb* jars. I don't think > 2.0.x will work at all with Java 6 update 4. Maybe if you also > endorse the jaxws-api jar. > > > CXF 2.1 will support Java 6 update 4 (but not updates 1-3). > > Dan > > On Wednesday 19 March 2008, Michael Berkowitz wrote: > > I've put this into Jira; I apologize if that wasn't the proper > > procedure. I did search quite a bit for any references to such a > > problem beforehand - no luck. > > > > Anyway, what I would expect is to get an exception from the invoke() > > that contains whatever information is in the fault message. This is > > killing my tester, who can never tell what's gone wrong without > > searching the Tomcat log for the corresponding server error. > > Sometimes even that doesn't show up, but that's for another email... > > > > If anyone can suggest what may be going wrong, I'm all ears. TIA. > > > > Michael -- J. Daniel Kulp Principal Engineer, IONA [EMAIL PROTECTED] http://www.dankulp.com/blog
