[Top] [All Lists]

Re: RFC2821bis-01 Issue 5: Syntax of ID clause in Received trace header

2007-03-23 03:05:47

In message <8A462C7931F053236D5EA172(_at_)as-s2n(_dot_)ietf68(_dot_)org>, John C Klensin <klensin(_at_)jck(_dot_)com> writes

In Section 4.4 on Trace Information, the "ID" clause of the Received field is defined as

      ID = CFWS "ID" FWS String / msg-id

This is claimed to be inconsistent with RFC 2822, which only permits msg-id.

Not so, RFC2822 has

received        =       "Received:" name-val-list ";" date-time CRLF

name-val-list   =       [CFWS] [name-val-pair *(CFWS name-val-pair)]

name-val-pair   =       item-name CFWS item-value

item-name       =       ALPHA *(["-"] (ALPHA / DIGIT))

item-value      =       1*angle-addr / addr-spec /
                        atom / domain / msg-id

So in RFC2822 an item-value can be an atom, which covers the cases given elsewhere by Bill McQuillan in this thread

id 1HUa8g-0004Bl-F8
id l2N38TpF018366
id l2N38UDd018377
id l2N38UN1018378
id F353995479
id A188A95316
id e3Zs1W01e1ZqGfk0000000
id <20070323033355(_dot_)LYDA18462(_dot_)fed1rmmtai109(_dot_)cox(_dot_)net(_at_)fed1rmimpi01(_dot_)cox(_dot_)net>

"String" is there to provide additional flexibility for gateways, etc., for which a message-ID format may not be appropriate, but it has been suggested that we remove it in the interest of consistency.

No need to remove it, as it is already consistent with RFC2822.

Question: Is the production correct as is, or should "String" be eliminated as a possibility?

"String" should be retained for consistency with RFC2822 and compatibility with current usage.

Note that, if the answer is "ok as is", we may want to think about whether we are happy with the syntax in in 2822.


Paul Overell         Internet Platform Development Manager, Thus plc