> It's not clear to me what exactely sender_verify / receiver_verify do.
Verify that the sender (and I assume receiver) is deliverable.
This means checking that exim can find a way to deliver to the user.
The delivery channel will do "easy" checks to ensure that the address looks
at least semi-plausible.
For local deliveries, this means that the mailbox is valid.
For remote deliveries, this means checking that the host looks plausible,
e.g. that there is a suitable DNS entry.
> Let's go back to the real world, "@unifr.ch" is not a local domain in
This means it will check "is the DNS info for unifr.ch plausible ?".
> # /usr/local/exim/bin/exim -bv bruno.vuilleminzzzzzzzzzz@???
No problem -- the *host* is valid -- it can't check the remote mailbox.
> Hmmm this looks like the verification looked for the mail exchanger
> for the domain and no more.
Yes -- it can't (in general) do any more.
> I am rigth or I am missing something?
No -- it verifies that the address as processible -- it cannot check that
remote addresses are valid.
> I was probably a bit naive and thinking that either a VRFY or
> a fake mail transmission was initiated on the other way.
Indeed.
And it wouyldn't actually prove anything -- it may just get to a relay.
> It's true that such a control may be considered "paranoid"
> and that checking the MX for a domain is enough.
Indeed.
--
* This is sent by the exim-users mailing list. To unsubscribe send a
mail with subject "unsubscribe" to exim-users-request@???
* Exim information can be found at http://www.exim.org/
