On 28/3/12 2:09 AM, Damien Saucez wrote:
I complete agree that using DNS protocol is not the best idea, even if we proposed it in the academic paper (ref given by Olivier). First, as Dino and Olivier say, DNS as it is today, is not very good at "boundary less" prefix matching (binary encoded names). Second, because the name is of limited size in number of characters you might have troubles with long addresses with instance ID because of the way you have to encode. And the final two points, which have been pointed out by Isidor, is that (i) negative mappings are not implementable as-is in DNS and more importantly (ii) DNS caching is poor for the LISP usage as it is caches the information with the exact matching and not the wildcard reply. However, I let Isidor extending the discussion on these two points as his the guy that discovered that (so I give to Cesar what belongs to Cesar!)
Olivier and Damien have accurately summed up the findings of the DNS based mapping system prototype we worked on.
- When matching a delegation hole the DNS NXDOMAIN answer does not include information on the hole that was matched in the zone file. This information is required by the resolver to return the negative map-reply to the ITR.
- The DNS answer contains the full name that was queried as well as the SOA prefix. It does not contain the name of the wild-card entry in the zone file that was matched by the query and that points to the map-servers. That entry represents the prefix being delegated to the map-server and is what ideally would be cached by the resolvers.
- Encoding XEID prefixes would be best done with the DNS bitlabel extensions that have been deprecated and removed from some (most?) implementations.
There are ways around all of the above but not without modifying the DNS server / resolver code.
Isidor _______________________________________________ lisp mailing list [email protected] https://www.ietf.org/mailman/listinfo/lisp
