I'm interested in the behaviour of callouts to verify recipients
( or "callforwards" as they were once known )
The manual implies exim uses the actual sender address of the
message, rather than just MAIL FROM: <>
This from 4.14 spec.txt - the last sentence is the relevant one:
------------------------------------------------------------------------
37.10 Address verification
<snip>
For a sender address, it behaves as if transmitting a bounce message and
sends:
HELO <primary host name>
MAIL FROM:<>
RCPT TO:<the address to be tested>
QUIT
For a recipient address, the MAIL command contains the sender address
of the message.
------------------------------------------------------------------------
However, on my system (exim4.14, solaris8) , the recipient callout simply
does MAIL FROM: <> just like it does when verifying senders.
Thinking further, this is perhaps just as well. Suppose the target of the
callout rejects the recipient perhaps because it has blacklisted the
enevelope sender presented in MAIL FROM:
Then, exim will presumably cache the fact this recipient is undeliverable
and thus reject future valid senders without trying a new callout.
So it would seem safer to verify recipients with MAIL FROM: <>
In other words, the code seems right, the manual seems wrong.
Unless of course I'm missing something !
--
Chris Edwards, Glasgow University Computing Service
