> On Oct 19, 2017, at 5:19 AM, Charlie Elgholm <char...@brightly.se> wrote: > > Yep, and that's probably why it doesn't work the way I want. I already > have some special domains listed in my retry-config, but Exim still > processes ALL the mail for that domain. Let's say I have 1000 mails to > hotmail, and get a "please go away"-421 message* 100 mails in, Exim > still tries the remaining 900 mails before they're put up for the > retry-config. Then when they're retried, well, 100 mails in I get the > same message, but Exim tries the remaining 800.. If I was the > rate-limiting server I would think that the connecting server is a bit > retarded ;)
To avoid this, it is best to stay under their rate limit on initial transmission, not just when running deferred messages. Of course if at all possible ask for and get from their postmaster a more generous rate limit. You could perhaps also special-case "421", that should never be a message-specific response. Indeed, even without special-casing 421, typically after 421 the server disconnects unilaterally, and the next SMTP command on your side sees a lost connection. If this is the final MX host for the destination, it should cause a "network failure" and the destination should be throttled after sufficiently many such failures. Don't know whether this is user-configurable, or requires code changes in Exim. -- Viktor. P.S. Yes, indeed be careful about hardcoding MX equivalence for domains you don't control. You'd have to monitor such an assumption closely, or get by without such mappings. -- ## List details at https://lists.exim.org/mailman/listinfo/exim-users ## Exim details at http://www.exim.org/ ## Please use the Wiki with this list - http://wiki.exim.org/