Re: [Exim] Brain Dead ISP's?

> > If the content is objectionable, it is _MUCH_ better to reject it in
> > the SMTP transaction rather than accept it and generate a bounce (or
> > some other form of non-delivery notice) later, because as often as
> > not the return-path is forged anyway.
>
> Be warned, however, that rejecting at any stage than just after RCPT TO tends
> to make at least some broken MTAs interpret this as a temporary / 4xx
> failure, and then they start to pound multiple times on your box - several
> times a minute, quite often.

It's this reason that I wish Exim had a database that it could store
sender/recipient cominations (possibly even hostip) and reject messages
earlier to future attempts.

--
Lab tests show that use of micro$oft causes cancer in lab animals