Hi Amin,
Thanks for your help. I believe you're explaining the inclusion proof, i.e. the certificate sent from the domin proves that it's in the log (that it is a valid certificate), not the merkle consistency proof, a proof the log server sends to browser.
Everytime a log server tells a browser here's a new STH (signed tree head) for a longer log (for example, by adding new certificates as my original question describes)Ā the brower will not accept the new STH until the log server has proved the new STH describes a suffix of the old STH, which the browser has already cashed and used.Ā
So my question is that, if the log server that contains the new certificate inserted by the CA is not the server the browser has previoulsy talked to, in other words, they don't share the same old STH, in case like this, the server seems unlikely to provide the browser a Merkel consistensy proof, and will it be rejected by the browser?
Please see:Ā
https://research.swtch.com/tlog (Section "Verifing a Log) andĀ
https://www.youtube.com/watch?v=UKdLJ7-0iFM&list=PLrw6a1wE39_tb2fErI4-WkMbsvGQk9_UB&index=18 (sencond half part).
Please point out if I don't make myself clear.
BR,
Kate