>-----Message d'origine----- >De : [EMAIL PROTECTED] >[mailto:[EMAIL PROTECTED] la part de Bart Mortelmans >Envoy=E9 : mardi 20 f=E9vrier 2007 12:17 >=C0 : [email protected] >Objet : [xmail] Re: Nolisting > > > >> If the primary MX is accepting connections and telling it to try >> again later, don't SMTAs try the primary again before sending to a >> secondary, hence the need for the RST packet in the firewall in the >> nolisting info ? I might be totally wrong about that, but I was >> thinking the secondary was only used when the primary was = unreachable >> or returning fatal errors ? >> =20 > >When an MTA get a fatal error (5xx) it normally shouldn't try again at =
>all, also not on the secondary MX. A fatal error can for example by=20 >"recipient unknown". Absolutly, fatal error codes 5xx (that means there was a successfull connexion to remote mx) have the effect to stop any try, and return a = ndr to the sender. > >I don't know if there is a difference between the=20 >retry-schedule for no=20 >response from the primary MX, or a temporary error (4xx). I=20 >would assume=20 >that most MTA's would handle both the same way, but haven't=20 >tested this yet. In theory, rfc don't say anything, so it make sense that a 'retry' is = not to be considered as a 'retry' except for the retry count, so the server = have to redo all tries as if the mail was just send first time (redo mx lookup, = mx test one by one, ...) No connexion to an mx IS NOT to be count as a retry, and each mx try is = in the same 'atomic' send > >Sincerely, >Bart Mortelmans > >- >To unsubscribe from this list: send the line "unsubscribe xmail" in >the body of a message to [EMAIL PROTECTED] >For general help: send the line "help" in the body of a message to >[EMAIL PROTECTED] > - To unsubscribe from this list: send the line "unsubscribe xmail" in the body of a message to [EMAIL PROTECTED] For general help: send the line "help" in the body of a message to [EMAIL PROTECTED]
