ietf-mxcomp
[Top] [All Lists]

Re: Backward compatibility with deployed SPF records (and choice of domain)

2004-06-26 01:57:44

On 6/26/2004 12:27 AM, Greg Connor wrote:

Does anyone have a strong objection to using unprefixed txt records?

--"Eric A. Hall" <ehall(_at_)ehsco(_dot_)com> wrote:
If the migration to a new RR requires queries for ALL (NOTE: this should
be avoided if possible), then it will be easier to implement if there is a
prefix. TXT and new RRs can both be fetched with a single query without
worrying about overload conditions from SOA, NS, MX, etc.

As long as the magic label is short enough, then there's really no reason
not to do it.


Hmm, I'm not sure if that was an objection? :)

Anyway, I thought one of Ed's points in his presentation was that adding a _prefix had its own set of problems, and wasn't recommended... Or maybe it was just wildcards that had the problem, but as PHB pointed out, you can define a wildcard and the prefixed and non-prefixed versions will both work. You can either have fine-grained control, or wildcards, but not both.

I don't see a strong technical reason why not, but I also don't see a strong reason reason we need it. I think I agree with you that ANY-type queries should be avoided.

The reason _spf prefix was dropped from SPF was NON-technical, but still an issue we should consider... people didn't understand it, and some still believed underscore was illegal, etc.

--
Greg Connor <gconnor(_at_)nekodojo(_dot_)org>