On 27/04/2018 21:15, Amelia Andersdotter wrote:
> On 2018-04-27 04:00, Brian E Carpenter wrote:

....
> i would have been slightly less annoyed had this not been the case. For
> this reason:
> 
>> This is not an area where anybody in authority gives a fig about what
>> the IETF says.
> 
> This is not reflective of my experience. The details are tedious, but
> RFC6302 in its current form, 

We need to look at one of those details. RFC6302 starts out by saying:

  "It is RECOMMENDED as best current practice that Internet-facing
   servers logging incoming IP addresses from inbound IP traffic also
   log:

   o  The source port number..."

Therefore, the whole recommendation applies *only* to servers that
happen to log incoming IP addresses. In effect, the document says:

IF you operate an Internet facing server AND you log incoming IP addresses
THEN you should also log the source port numbers (etc.).

That is a purely technical statement, because with address sharing
in use, there is no point in logging addresses without ports.

The document does *not* say:

IF you operate an Internet facing server
THEN you should log incoming IP addresses, source port numbers (etc.).

That would be a completely inappropriate thing for the IETF to say,
because it's outside our technical remit.

If draft-daveor-cgn-logging was adopted as an IETF document, it
should IMHO be subject to the same test: is it describing how to
do something correctly, if you're doing it at all? At least some
of the language in section 7 would need tuning for that. Other
parts seem OK, like "In cases where a software package has
support for logging of incoming source port,..."

Regards
   Brian

_______________________________________________
Int-area mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/int-area

Reply via email to