Robert Schetterer a écrit : > Am 23.03.2010 00:14, schrieb mouss: >> Mauro Faccenda a écrit : >>> Hi Reinaldo, >>> >>> On Fri, Mar 19, 2010 at 7:34 PM, Reinaldo de Carvalho >>> <reinal...@gmail.com> wrote: >>>> On Fri, Mar 19, 2010 at 6:40 PM, Mauro Faccenda <facce...@gmail.com> wrote: >>>>> Alternative to that patch? I did some searches and as far as I could >>>>> see, none of the alternatives (that I've found) could reject the >>>>> message in SMTP. I think it's nice to not generate bouncing >>>>> messages/backscatter. But if someone have an alternative besides that, >>>>> it's always welcome. >>>>> >>>>> >>>> 1. http://postfixquotareject.ramattack.net/ >>> I took a look at it and, unless I missunderstood, I'll need to add a >>> line for each mailbox in the database and that's inviable for my >>> setup. >>> >>> Also I droped the use of the VDA patches, since it implements >>> everything in Postfix's LDA and I am actually using the Dovecot's >>> (deliver). >> well, if you use dovecot, then forget about vda and the like. >> >> if you really need quota, then dovecot-sieve has quota support. if >> you're not happy (now you no longer play the game?), then report what's >> not good enough. > > overquota dovecot is not done at smtp income stage > a overquta bounce of deliver might get backscatter >
No, it's not backscatter. if bounces happen too often, then there is a serious problem (bounces are only part of the problem). but if they occur rarely, then that's not a problem. after all, - mailing lists do send confirmation requests to possibly forged addresses. nobody has said this is backscatter. - unless you go the fanatic way that says you must deliver to the mailbox before sending "ok", delivery problems may occur. in which case, bounces are desirable. back to the quota issue, this was already discussed here multiple times. there are many mitigation approaches, and I won't repeat the discussion. one suggested approach is to update an access map.
