Hello Phil and the others,
thank you for responding.
Philip Hazel <ph10@???> (Fr 12 Jan 2007 10:53:03 CET):
> On Thu, 11 Jan 2007, Heiko Schlittermann wrote:
>
> > in my RCPT ACL I've:
> > ...
> > require message = Huh (sender)!
> > verify = sender/callout=random
> >
> > require message = Huh (recipient)!
> > verify = recipient/callout=random
> > ...
> >
> > I expected exactly my message in the server response, nothing but my
> > message. But in the following examples you'll see that it's true only
> > for the recipient verification.
>
> If you want just your message for the sender verification failure, then
> you must do the sender verification at MAIL time, not at RCPT time.
Hm. But this conflicts with my attempt to accept *all* mails directed to
postmaster. (Yes, I could use some variable and delay the rejection, but
this would be ugly.)
>
> > Now testing it with BAD SENDER:
> >
> > # exiacl -f pitti@??? -t info@???
> > **> /usr/sbin/exim -C /etc/exim4/exim4.conf -oMi 145.253.107.250 -bhc 172.20.1.8
> > LOG: no host name found for IP address 172.20.1.8
> > < 220 paff.bioz.tzdresden.de ESMTP Exim 4.63 Thu, 11 Jan 2007 22:40:52 +0100
> > > EHLO schlittermann.de
> > < 250-paff.bioz.tzdresden.de Hello schlittermann.de [172.20.1.8]
> > < 250-SIZE 52428800
> > < 250-PIPELINING
> > < 250-STARTTLS
> > < 250 HELP
> > > MAIL FROM: pitti@???
> > < 250 OK
> > > RCPT TO: info@???
> > LOG: H=(schlittermann.de) [172.20.1.8] sender verify fail for <pitti@???>
> > LOG: H=(schlittermann.de) [172.20.1.8] F=<pitti@???> rejected RCPT info@???: \
> > Sender verify failed
> > < 550-Verification failed for <pitti@???>
> > < 550-Previous (cached) callout verification failure
> > < 550 Huh (sender)!
>
> The reason for this is to make it clear that it is the sender that has
> failed to verify, and to state precisely which email address failed.
> Otherwise it could be very confusing if you had something like:
>
> RCPT TO:<x@@y>
> 550 Verification failed
Yes, this I understand. But I'd suppose that I know what I'm doing if
I use my own message. And always I can use $acl_verify_message if I
want to expose the real reason.
Somehow I feel the current beheviour not consistent. Is there any
chance to change it?
Best regards from Dresden
Heiko Schlittermann
--
SCHLITTERMANN.de ---------------------------- internet & unix support -
Heiko Schlittermann HS12-RIPE -----------------------------------------
gnupg encrypted messages are welcome - key ID: 48D0359B ---------------
gnupg fingerprint: 3061 CFBF 2D88 F034 E8D2 7E92 EE4E AC98 48D0 359B -
