Stephane - Thanks for your review of DNS Map! I also very much enjoyed our face-to-face discussion in Philly. It's good to have you in RRG.
After reading your document, I fail to understand how the DNS query is constructed, when the "key" is an EID (if it is an host name, there is no problem).
If the key is an EID, the query will be for reverse DNS. Does this answer your question?
You say that the key of the lookup can be a single address or a prefix. In the latter case, how do you encode it in the query?
The lookup key is always for a specific host, i.e., either a domain name or an address. The MAP resource records themselves may contain aggregated information, however. They include (in the RDATA field) a pair of edge address prefix and transit address prefix -- although prefix lengths may be 32 bits (IPv4) or 128 bits (IPv6) to enable non-aggregated, per-host mappings. A DNS query for a particular edge address then returns all MAP resource records for which the edge address prefix includes the edge address from the query. The owner name of the returned records is set to the lookup key from the query. Depending on the DNS server implementation, the owner name may be filled in dynamically given the lookup key in a DNS query, or MAP resource records must be replicated for each possible owner name. The latter can typically be accomplished via macros, which many DNS servers provide to generate resource records for a range of owner names.
Figure 1 in your document did not help. Network A wants to contact EID ABC::1. But the figure shows its ITR performing a query for 1000::1... The figure seems to contradict the text.
Thanks for catching this. The ITR should perform a query for "ABC::1" instead of "1000::1". I am going to revise the document based on your review. - Christian -- to unsubscribe send a message to [EMAIL PROTECTED] with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg
