On 21/02/2017 01:33, John C Klensin wrote:
>
>
> --On Monday, February 20, 2017 10:07 PM +1100 Mark Andrews
> <marka@xxxxxxx> wrote:
>
>>
>> In message
>> <SG2PR06MB071061291C3DC252AA62FCB2C15E0@SG2PR06MB0710.apcprd06
>> .prod. outlook.com>, Danny Niu writes:
>>> Questions:
>>> Is "Berkerly Sockets API" defined seperately from the BSD
>>> manpages? Or is it just sections of the BSD manpages? What
>>> happened to "Berkerly Sockets API"?
>>>
>>> Proposal:
>>> Folks at POSIX are a bit unwilling to dis-certify some
>>> allegedly existing systems, and think it'd be better IETF
>>> note the purpose of port 0, so that existing app/sys woudn't
>>> break.
>>> So is it too soon to start drafting?
>>
>> Well UDP source port 0 means don't reply (RFC 768). It's for
>> uni directional streams.
>>
>> As for 0 to select a ephemeral port that is a BSD sockets
>> convention. That isn't something the IETF should specify.
>
> While _assignment_ of a por is an IETF matter and I mostly agree
> with Mark, recognition of how one is being used is is a little
> different.
>
> It seems to me that this is rather more an IANA registry matter
> than a standardization one
Not quite though. Noting the RFC 768 usage as a source port, aren't
all these uses equivalent to saying 'Not valid on the wire as a
destination port'? That sounds like something the IETF should say,
as a protocol matter, and it leaves the value open for use by APIs
or other software.
This rings bells for me. Only yesterday I had to fix an unassigned
variable bug in my code for draft-ietf-anima-grasp by adding this:
else:
listen_port = 0
Brian
> and that, given practices today, it
> would be reasonable to annotate the registry by adding "used
> for" or "known to be used for" to "reserved".
>
> That should be mostly a housekeeping matter: could one of the
> relevant ADs speak up and indicate whether they want an I-D, a
> note (perhaps just this thread), a formal liaison request from
> POSIX, or something else?
>
> john
>
> .
>