Is there a JIRA bug on that or some demonstrated code? I've never seen that behavior. I use the singleton pattern when I do things that could create conflict between threads, so if something jumps the lock that generally will stop it from causing mischief. -- Andy
>>> [EMAIL PROTECTED] 12/22/04 05:33PM >>> I have seen one threading related problem with Axis 1.1, If there are a number of threads that attempt to create the very first client stub in a given process concurrently, then I've seen NPE's from deep in the bowls of the AxisEngine class, have never had the time to pin down exactly what causes it. Cheers Simon > -----Original Message----- > From: ANDREW MICONE [mailto:[EMAIL PROTECTED] > Sent: Wednesday, December 22, 2004 4:30 PM > To: [EMAIL PROTECTED] > Subject: Re: Locator and Stub thread safe > > I don't know the formal answer but as far as I can tell the > answer is yes, the generated code is thread safe. I make use > of a number of J2EE's threading features in my generated > service stubs and I've not seen any conflicts...or I've been > extremely lucky. It's on a four-way SMP system, so I'd be > likely to trigger them. -- Andy > > >>> [EMAIL PROTECTED] 12/22/04 06:58AM >>> > If I use WSDL2Java (from Axis 1.1) to create a client, will > the generated Locator be thread safe? > What about the generated service Stub? > > > >
