Dear Stephane, Thanks a lot for your comments.
Please see inline. BRs Linda Stephane Bortzmeyer <bortzme...@nic.fr> 写于 2015-07-04 04:22:43: > Stephane Bortzmeyer <bortzme...@nic.fr> > 2015-07-04 04:22 > > 收件人 > > wang.c...@zte.com.cn, > > 抄送 > > suzworldw...@gmail.com, tjw.i...@gmail.com, > huang.sunli...@zte.com.cn, meng.w...@zte.com.cn, dnsop@ietf.org > > 主题 > > Re: New Version Notification for draft-huang-dnsop-synchronization- > resolver-server-00.txt > > On Fri, Jul 03, 2015 at 10:14:50AM +0800, > wang.c...@zte.com.cn <wang.c...@zte.com.cn> wrote > a message of 143 lines which said: > > > I have just submitted a new draft which tries to work out how to > > synchronize the RRs information between resolvers and DNS servers. > > I have two serious concerns with this draft. The most important one is > that I do not see an use case. DNS is and has always been "loosely > consistent". It was never envisioned that a change in the > authoritative name servers would be reflected "instantaneously" in > every resolver. If someone wants faster changes, he can always use > (reasonable) short TTLs. So, you really should add a section > explaining what is the use case. //Linda: Yes, I agree with you. Next version, we plan to add one or two use cases to highlight the issue proposed in this draft. Then we can continue our discussion whether the solution is OK. > > The second concern comes from the observation that authoritative name > servers are fast and very resilient because they don't keep > state. Asking them to memorize all their clients, in order to notify > them later of a change in the data, is a sure way to weaken them. //Linda: If we can confirm this issue is sure to exist. There should be a feasible solution to leverage it and then we can work out it. > Thanks a lot:)
_______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
