Re: [IPFIX] New Version Notification - draft-ietf-ipfix-flow-selection-tech-13.txt
Benoit Claise <[email protected]>
| Newsgroups | gmane.ietf.ipfix |
|---|---|
| Message-ID | <[email protected]> |
On 25/02/2013 10:28, Paul Aitken wrote:
> Benoit,
>
>> On 24/02/2013 17:40, Paul Aitken wrote:
>>> Benoit, Authors,
>>>
>>> Rather than defining a new "flowSelectorAlgorithm", shouldn't the
>>> proposed algorithms be added to the existing selectorAlgorithm #304?
>>>
>>> This only requires removing "packet" from the existing definition
>>> (or s/packet/packet and flow/), and allow that a Selection Process
>>> can be either a packet selection process or a flow selection process.
>> That would be nice, but I guess we would have to update all the PSAMP
>> RFCs. In particular, http://tools.ietf.org/html/rfc5477#section-8.2.1
>
> No, because we already decided that IANA's IPFIX registry is the
> master repository over the IPFIX RFCs - the registry can be updated,
> corrected, and extended, whereas the RFCs cannot.
Master registry for the IE definitions, granted.
However, IPFIX is not only composed of IPFIX IEs, but also of
specifications on how to use those IEs. For example, RFC 5476, RFC 6728.
So changing this definition seems very easy, but I'm concerned by the
implications on the current set of RFCs.
On top of that selectorAlgorithm is special because there is an IANA
registry bound to it (http://tools.ietf.org/html/rfc5477#section-8.2.1)
The selectorAlgorithm registry is maintained by IANA. New
assignments for the registry will be administered by IANA and are
subject to Expert Review [RFC5226 <http://tools.ietf.org/html/rfc5226>].
That's one basic required change, for which a new RFC is required.
Regards, Benoit
> So the same must hold for the PSAMP RFCs.
>
> P.
>
_______________________________________________
IPFIX mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ipfix