At 10:30 +0000 2/20/07, Jim Reid wrote:
No Ed, they're ambiguous. Paul's right. Consider two companies each of whom use 10/8 on their intranets. They set up a joint venture and interconnect their nets. Whose 10/8 is telling the truth for some 10/8 address and which one is lying? What goes into the DNS? Let's say they use a common 10/16 for the joint venture. If someone's in that /16, which (parts of) 10.in-addr.arpa do they resolve against for other parts of the network?
"Truth" and "lying" ? No, it's a conflict that needs to be part of the integration of the two companies. There are people whose job is to iron this stuff out. It's not as dramatic as "truth" and "lies." It's simply a cost of doing business.
If it's ambiguous that is distasteful, let's recommend a ban on putting anycasted IP addresses in DNS. Or ban multi-homed (as you can't really tell anycast and multi-homing apart) addresses too.
What I think we're trying to achieve here is define an element of the "standard way to properly use RFC 1918 space".
Which is what this thread has morphed into. Refining RFC 1918 is not in the charter of DNSOP however. How does this come back to DNS operations?
If the answer is to recommend that NS records don't point to (servers with) RFC 1918 space, you are recommending that anyone using RFC 1918 space go through hoops to get DNS working. Getting any consensus on this will be as easy as that document on "publishing unusable addresses." (Unusable to whom?)
-- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Edward Lewis +1-571-434-5468 NeuStar "Two years ago you said we had 5-7 years, now you are saying 3-5. What I need from you is a consistent story..." _______________________________________________ DNSOP mailing list [email protected] https://www1.ietf.org/mailman/listinfo/dnsop
