Dear colleagues, Over in dnssd, there's a draft on the agenda about long-lived queries. It's expired, but there are different options. For the mDNS case, it turned out that LLQ were perhaps not so important; but on the Internet, they're going to be.
See the agenda at http://tools.ietf.org/wg/dnssd/agenda. The long-expired draft is at http://www.ietf.org/archive/id/draft-sekar-dns-llq-01.txt. There's some proposal to do TCP for this, also. One of the interesting things about the LLQ approach is that it implies clients contacting authoritative servers directly and cutting out the intermediate resolvers. It seems to me this proposal has fairly big implications for the use of the protocol, and therefore it'd be good to get operational eyes on it. At the same time, it represents an intriguing way to handle cases where this might be important. So I think it's important that those interested take the idea seriously and not just dismiss it as unworkable in the general case. It might not be a general-case tool, but one that needs to be carefully circumscribed for certain uses (if that is even possible). Best regards, A -- Andrew Sullivan a...@anvilwalrusden.com _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
