RE: WSO2 -> Axis issues (PLEASE READ, SPEC/TEST ISSUES)

One way or another we should definitely add some tests. I agree that defaulting to anonymous on all WSA errors doesn't sound like the right approach.

Mark.



-----Original Message-----
From: Glen Daniels [mailto:gdaniels@sonicsoftware.com]
Sent: Wed 3/15/2006 12:01 PM
To: Mark Little
Cc: public-ws-addressing-tests@w3.org
Subject: RE: WSO2 -> Axis issues (PLEASE READ, SPEC/TEST ISSUES)
 

> I agree that's not what the spec says now, but I thought from 
> the minutes that there had been discussion around clarifying 
> it. Apparently not. Did anyone take an AI on Monday to try to 
> resolve this further?

No, and I don't think we plan to change anything else - that's why I
suggested putting in another explicit test for this behavior (ReplyTo ==
none, bad To/Action header, no fault returned).  It appears that all the
implementations except Axis are currently assuming "faults to anonymous
on any WSA error", which is, I believe, wrong with the status quo.

--Glen

> -----Original Message-----
> From: Glen Daniels [mailto:gdaniels@sonicsoftware.com]
> Sent: Wed 3/15/2006 11:14 AM
> To: Mark Little
> Cc: public-ws-addressing-tests@w3.org
> Subject: RE: WSO2 -> Axis issues (PLEASE READ, SPEC/TEST ISSUES)
> 
> 
> Hi Mark:
> 
> > Having caught up on my email backlog it seems that we need to
> > tighten up the specification in terms of what it says about
> > errors that occur during the processing of the WSA header. It
> > does appear from the discussion in the WG that the consensus
> > is that faults must be dealt with as though WSA was not being
> > used at all. I missed that part of the meeting on Monday, so
> > was there anything else said that isn't covered in the minutes?
> 
> I don't think that's quite correct.  IIRC, we simply clarified that if
> there are any problems with the headers, you cannot set values for the
> abstract properties associated with the particular bad headers.
> Therefore if there is a duplicated <To> for instance, normal <FaultTo>
> and <ReplyTo> processing would still occur (barring any problems with
> those headers of course), and faults would be delivered to 
> the <FaultTo>
> EPR if present, and the <ReplyTo> EPR otherwise.  If there were a
> duplicated <FaultTo> you wouldn't have a [fault endpoint] 
> property, and
> would therefore default to <ReplyTo> anyway.  That's my understanding.
> 
> I would certainly be fine from a design perspective to say that faults
> doing WSA processing are treated as if WSA is not in effect, 
> but that's
> not what the spec says now.
> 
> --Glen
> 
> 
> 
>