Re: [sipcore] Spencer Dawkins' Yes on draft-ietf-sipcore-dns-dual-stack-07: (with COMMENT)

[Spencer Dawkins at IETF <[e-mail]>]

Hi, Dale,

On Fri, Aug 19, 2016 at 2:13 PM, Dale R. Worley <[e-mail]> wrote:

> "Spencer Dawkins" <[e-mail]> writes:
> > Could you provide some explanation for the reader as to why SIP is
> > "different", following this statement? Even an example would be useful.
> >
> >    Unfortunately, in common SIP situations, it is not possible to "race"
> >    simultaneous request attempts using two address families.
>
> That's a good point, because the reason is non-obvious to a reader who
> hasn't hassled with SIP in unreliable network environments.
>
> Racing in HTTP is the sending of simultaneous TCP SYNs to two different
> addresses.  The client then continues the handshake and sends the
> request to the address that responds first.  The crucial fact is that
> opening a TCP connection to an HTTP server alone does not change the
> server's state.
>
> SIP requests are usually transmitted as single UDP packets, so sending
> two copies of the request to two different addresses risks having two
> copies of the request propagating through the SIP network at the same
> time -- the difference from HTTP is that the SIP element cannot test the
> route in a non-state-changing way.
>
> If two copies of the same request arrive at one client, the client must
> reject the second of them with a 482 response.  But this rule is not
> sufficient to prevent user-visible differences in behavior because a
> proxy that is upstream of the second request to arrive at the client may
> (almost immediately!) serially fork the second request to further
> destinations.  A version of this scenario that I saw in the wild is
> diagrammed in
> https://www.ietf.org/mail-archive/web/sipcore/current/msg06853.html.
>
> I've added this explanation to section 3.2.  You can see the change in
>
> https://www.ietf.org/rfcdiff?url1=draft-ietf-sipcore-dns-
> dual-stack-07&url2=http://svn.resiprocate.org/rep/ietf-
> drafts/worley/draft-ietf-sipcore-dns-dual-stack-08.txt
>
> Dale
>

That's a terrific response to my comment. Thank you for including your
response in the document!

Spencer