On Tue, Apr 18, 2017 at 1:28 AM, Hugo Landau <[email protected]> wrote:
> > Hey Hugo, > > Thanks for the PR. Jacob and I chatted a bit about this, and wanted > to > > propose a slightly different approach: > > [1]https://github.com/ietf-wg-acme/acme/pull/297 > > What do you think? Others on the list have any thoughts? > > This looks mostly OK. > > My only concern is that it's not clear how a client can determine when a > server is still attempting automatic retries. Is this to be detected via > the absence of the Retry-After header? Something more explicit might be > nicer (and less susceptible to misimplementation). > The idea is that as long as the challenge status is still "pending", the server is still retrying. Once it gives up, it marks the challenge as "invalid". > Also, is each automatic retry to result in an error being appended if > unsuccessful? This could get quite spammy depending on how many times a > server repeats verification. Possibly amendment of the problem document > format to add a repeat count (in the spirit of "last message repeated N > times") might be desirable. Or that might be my usual overengineering. > Yeah, I'm not to worried about spamminess here. The server can always limit the number of retries. --Richard > > Hugo Landau >
_______________________________________________ Acme mailing list [email protected] https://www.ietf.org/mailman/listinfo/acme
