> > i'm planning to submit a draft recommending that "id" be bidirectional > > (so, both clientid and serverid) with the client's id being the > > solicitation of the server's id. this doesn't require rev'ing the EDNS > > version number > > but will that still require more than one query?
no. it uses the presence of an ID option inside the EDNS OPT RR in the request to signal the desire to see one in the response. > As I see it, the whole idea of this draft and the one from Rob (which > implements the EDNS tweaking you mention), is to retrieve this > information with a single query. yes, that's still the goal. and to rob's observation that the need for this in clients isn't symmetric to the need for it in servers, i think we can answer that by making a small change to the serverid and/or nsid drafts making it symmetrical, saying that no client-side options are defined at this time, and making the server's response into an attribute/value list. -- Paul Vixie . dnsop resources:_____________________________________________________ web user interface: http://darkwing.uoregon.edu/~llynch/dnsop.html mhonarc archive: http://darkwing.uoregon.edu/~llynch/dnsop/index.html
