RE: WebDAV Bindings - Issue Yaron.507

My point is that the binding could fail not due to cross-server anything but
an intra-server something. That is why I brought up the FrontPage example.
If you try to bind between volumes it will fail. Yet given the current
definition FrontPage could not use a 507 in that case. That just seems
broken. That is why I wanted to widen the language so it could cover the
cross-volume scenario.

		Yaron

> -----Original Message-----
> From: Slein, Judith A [mailto:JSlein@crt.xerox.com]
> Sent: Wednesday, January 19, 2000 5:49 PM
> To: 'ccjason@us.ibm.com'; Yaron Goland
> Cc: w3c-dist-auth@w3.org
> Subject: RE: WebDAV Bindings - Issue Yaron.507
> 
> 
> I don't have a strong commitment to 507 either, but for what 
> it's worth the
> rationale was that any creation of a cross-server binding requires
> out-of-band collaboration between the servers.  So it seems 
> very likely that
> a lot of servers will fail requests to create a binding to a 
> resource on
> another server, so it seems useful to have an error code for 
> this case.
> 
> --Judy
> 
> > -----Original Message-----
> > From: ccjason@us.ibm.com [mailto:ccjason@us.ibm.com]
> > Sent: Tuesday, January 18, 2000 8:40 PM
> > To: Yaron Goland
> > Cc: w3c-dist-auth@w3.org
> > Subject: Re: WebDAV Bindings - Issue Yaron.507
> > 
> > 
> > 
> > >>
> > In section 5.5 the 507 error code is written as "507 
> > (Cross-Server Binding
> > Forbidden): The server is unable to create the requested 
> > binding because it
> > would bind a segment in a collection on one server to a 
> resource on a
> > different server."
> > 
> >         What does a server have to do with anything? If you 
> > try to bind two
> > resources in different volumes on a FrontPage server the 
> > server will have
> > to fail the BIND even though the resources are on the same 
> server. In
> > general bringing in the server is almost always a bad idea 
> > since resources
> > can be spread out all over the place and the reasons for 
> > various failures
> > may or may not have anything to do with how those resources 
> > are laid out on
> > the servers. As such I move that the language for the 507 
> > error code be
> > altered to read that the resource was unable to create a 
> binding to a
> > destination and to leave the matter at that. All mentions 
> of the word
> > server should be stricken.
> > >>
> > 
> > Hmmm.  I don't have a strong preference on whether we should 
> > create a new
> > status code for lack of support for remote connections.  At 
> > some point we
> > might find we need one.   Anyway.... the status code that 
> > you're suggesting
> > doesn't seem to suggest anything except that the server can't do it.
> > Can't we just use 500 for that?   And if so, shouldn't we 
> > mention 500 it in
> > the spec?  Or is 500 too obvious to mention?
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> 

Received on Saturday, 22 January 2000 05:32:49 UTC