Mark Delany writes: > True, there is no mandate, but all implementations that I'm aware of, > have implemented a white list. While the language is softer in -02, is > it necessary at all as it will only continue to encourage white list > behavior just as the previous language did.
The reality is that at least one major provider of authoritative ECS service has realistically no chance at all of removing its whitelist in the visible future. At least one major recursive implementer has found that they must use a whitelist for sending ECS as well or they pay a significant penalty on first hit to a server that mishandles it. That is not to say that I disagree with your basic point. Whitelists are a pain to accurately maintain, of that there is no doubt. They are, however, also reality. > IOWs, why not remove 11.2 completely? In my opinion, it is important to make the reader aware that such whitelisting does exist and that ECS won't just work out of the box between any two ECS-supporting recursive/authoritative instances. _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
