[ppml] CIDR support for whois.arin.net / merging whois with rr data?
sandy at tislabs.com
sandy at tislabs.com
Fri Sep 15 08:12:36 EDT 2006
- Previous message: [ppml] CIDR support for whois.arin.net / merging whois with rr data?
- Next message: [ppml] RETRACTION OF Policy Proposal 2006-6: Bulk WHOIS agreement expiration
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Bill Woodcock said: >The IRR whois contains only non-authoritative data, either submitted >directly, submitted through third parties, mirrored from sites which >received it directly, or mirrored from sites which received it from third >parties. > >If one's interested in the chain of responsibility backing up the veracity >of the data, these things are of some interest. Three somewhat orthogonal comments: The directly submitted data for resources for which ARIN is authoritative can be validated. And as I understand it, ARIN *does* check for number resource objects being registered to be sure that the registrant is the same as the POC for the corresponding whois record. So some authorization is possible. As I understand it, the route objects are not subjected to the same validity check, although it should be possible for ARIN to check that the registrant is the same as the POC for the corresponding whois record for the address. There's an RFC that talks about security for rpsl databases and objects (RFC2725, RPSS) that says a route object should be validated against BOTH the address and AS maintainers. A I understand it, RIPE implements this check in their database. If ARIN does validate (some) submissions, but receives other submissions from other sources, it seems worthwhile to flag that difference in the object. Something to say whether or not ARIN validated the source. --Sandy
- Previous message: [ppml] CIDR support for whois.arin.net / merging whois with rr data?
- Next message: [ppml] RETRACTION OF Policy Proposal 2006-6: Bulk WHOIS agreement expiration
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the PPML mailing list