Change Details

As discussed yesterday in -traffic, we must take into account certain rate limits imposed by LE to avoid reaching them ourselves and kind of DoS-ing the certcentral service. - [] 5 failed validations per account, per hostname, per hour --> certcentral should be able to delay the retries issuing a certificate which validations are failing systematically - [] 5 duplicates per week (exact same CN/SAN list, account/crypto doesn't matter). --> certcentral should be able to delay the retries iff the certificate has been issued by LE but for some reason certcentral fails to persist on disk (host runs out of disk, issues on certificate serialization...)

As discussed yesterday in -traffic, we must take into account certain rate limits imposed by LE to avoid reaching them ourselves and kind of DoS-ing the certcentral service. - [x] 5 failed validations per account, per hostname, per hour --> certcentral should be able to delay the retries issuing a certificate which validations are failing systematically - [] 5 duplicates per week (exact same CN/SAN list, account/crypto doesn't matter). --> certcentral should be able to delay the retries iff the certificate has been issued by LE but for some reason certcentral fails to persist on disk (host runs out of disk, issues on certificate serialization...)

As discussed yesterday in -traffic, we must take into account certain rate limits imposed by LE to avoid reaching them ourselves and kind of DoS-ing the certcentral service. - [x] 5 failed validations per account, per hostname, per hour --> certcentral should be able to delay the retries issuing a certificate which validations are failing systematically - [] 5 duplicates per week (exact same CN/SAN list, account/crypto doesn't matter). --> certcentral should be able to delay the retries iff the certificate has been issued by LE but for some reason certcentral fails to persist on disk (host runs out of disk, issues on certificate serialization...)