Re: Last Call: draft-dusseault-http-patch (PATCH Method for HTTP) to Proposed Standard

[Eliot Lear <lear at cisco.com>]

On 10/30/09 10:27 PM, Nikunj R. Mehta wrote:
> On Oct 30, 2009, at 11:01 AM, Lisa Dusseault wrote:
>
> > Hi,
> >
> > On Wed, Oct 28, 2009 at 11:03 AM, Nikunj R. Mehta
> > <nikunj.mehta at oracle.com> wrote:
> > > This draft places unreasonable restriction on servers about processing
> > > requests. Specifically, in §2.2,
> > >
> > > [[
> > > Concurrent modification: When a server receives multiple concurrent 
> > > requests
> > > to modify a resource, those requests SHOULD be queued and processed 
> > > in the
> > > order in which they are received. If a server is incapable of queuing
> > > concurrent requests, all subsequent requests SHOULD be rejected with 
> > > a 409
> > > (Conflict) until the first modification request is complete.
> > > ]]
> > >
> > > RFC2616 describes the above status code (409) but not in the context 
> > > of a
> > > particular type of HTTP request. I fail to see why this draft has 
> > > mandated
> > > specific error codes and specific server behavior in response to 
> > > certain
> > > requests. It curtails server behavior without a good reason.

I won't speak to a specific error code, but it's fairly clear that what 
the authors are attempting to do is provide for a minimum of chaos when 
concurrent PATCHes are requested.  These operations are meant to be 
atomic in spirit if not in word.  You can't have two atomic operations 
on the same object at the same time.

Eliot