Re: [IPFIX] IPFIX IE Doctors
Brian Trammell <[email protected]>
| Newsgroups | gmane.ietf.ipfix |
|---|---|
| Message-ID | <[email protected]> |
Hi, Paul, Thanks for the comment. I agree in principle; however I'd like to suggest something that doesn't have the risk of mixing information element status (defined in 5102/5102bis) which provides information to IPFIX implementors, and this, which provides information to people extending IPFIX (who are often themselves implementors of the new application, but this is still subtly different)... Since the current/deprecated/obsolete status may have runtime impact (i.e. logging at the collector "hey your exporter is ancient"), while "please don't define new IEs based on this one" is meant for humans, I suggest that we update each Information Element with a note in the Description field, stating "This Information Element should not be used as an example or precedent when defining new Information Elements," or similar. (If we ensure the string is easily searchable, we can make things easier for IE Doctors.) Thoughts? Cheers, Brian On Nov 17, 2011, at 9:09 AM, Paul Aitken wrote: > Brian, > > SInce IANA is _the_ repository, that's the place where people will be copying IPFIX IEs from. > > So that's the place where the "not recommended" status should be recorded. > > It shouldn't be in a separate wiki, nor in an RFC, since we know that people are lazy and don't look in multiple places. > > The status could be "not recommended", "discouraged" or "disapproved". > > Or convince the originator of how to export the information correctly, and mark the IE as deprecated. > > P. _______________________________________________ IPFIX mailing list [email protected] https://www.ietf.org/mailman/listinfo/ipfix