On Fri, 19 Apr 2013, Steve Crocker wrote:
If someone is paranoid about the possibility of being spoofed, he can compare the results from multiple publishers and/or rotate among the many publishers. But there's no need for the publishers to coordinate among themselves, except for the standard format, and there's no need for a formal quorum of witnesses. (I guess if someone wanted to advocate a best practice of using a quorum of witnesses, that's ok with me, but I view that as an added layer, not necessarily required.)
You have solved the easy human problem. Now do the same for fully automated computers that are turned on after 10 years. There is currently one publisher (ICANN pem bundle plus static web page with certs signed by particular CA). Will that page be there in 10 years? Where will the other publications be? A few humans can figure this thing out easilly, and adapt. The problem of rolling the root key is automation. Imagine your previous DSL router was doing DNSSEC with the previous root key. Your current DSL modem dies, and you power on your old one. What will happen? Now repeat, and say that the SHA2 family was broken and SHA4 is now the standard. Your old modem talks SHA2 only. I don't think we as a group have ignored this problem - we have just not found a proper solution yet. Rolling the root key now will be pretty meaningless as a test for deployment issues in the future. And if the root zone operator wants to test their procedures, they can do so in a lab. Paul _______________________________________________ DNSOP mailing list [email protected] https://www.ietf.org/mailman/listinfo/dnsop
