Sanjiva Weerawarana wrote:
On Thu, 2007-02-08 at 16:35 +1300, Dennis Sosnoski wrote:
+1
This is a capability my clients have often wanted. I'm not sure I see
the need to do it at the MessageReceiver level, though. The approach
I've thought about is to use handlers. The handlers could not only
The problem is that pinging at the handler level basically amounts to
just asking "is there such a service?".
That's true if the handler is not providing any useful response
information. If the handler can report that the service is accepting
requests and delivering non-Fault responses that seems to be more useful
information than you'd get just by going to the message receiver.
Having it go to the message
receiver allows the ping to be responded to based on more useful data-
such as did the impl class or script or whatever load etc.. So its
really a question of fidelity- yes some level of liveness checking can
be done at a handler level but you get much improved info by getting the
ping to the message receiver.
Only if the user actually implements the ping-handling code themselves,
and checks the various conditions that can go wrong in the operation of
the service code. Was that the intention, that this be exposed to the
user to implement? I don't know see offhand how this would work, given
that somewhere around 95% of Axis2 services are using generated code.
Just having it go to a MessageReceiver default implementation seems to
add zero value beyond what you get by querying the admin service.
- Dennis
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
