On (2014-05-23 08:30 +0000), Ray Bellis wrote: > I've made him aware of the update to Mark's HTTP SRV draft, but there appears > to be a problem that SVCINFO addresses that SRV doesn't, namely declaration > (or discovery) of which version of the protocol is supported by the endpoint.
ACK. We have great window of opportunity with HTTP2 to mandate use of some RR which delivers all information client needs to make informed decision how to redundantly access web service. Then we can reduce client query to 1 and gain new services and reduce need for having single redundantly routed address for web server. -- ++ytti _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
