We have some users which utilize some forwarding features from gmail, which
is causing some odd F=<> fields being sent to our exim servers. These
munged up from addresses fail our sender lookup validation, and the email
gets bounced. All of this is explained and understood.
The munged up addresses have valid characters in them.
i.e.
F=<joe.blow+caf_=some=mydomain.ca@???<F=joe.blow+caf_=some=address.ca@???><joe.blow+caf_=some=mydomain.ca@???>
>
Where joe.blow is the sender username I want to validate, and "some" is the
destination address. The RCPT field of this message is correct, so I'm not
too worried about it. I need the F field to read F=<joe.blow@???>
What would be some ideas for a good rewrite rule that would capture this
kind of scenario. I still want to validate the sender but my understanding
is that the rewrite rules occur before any of the acl's or other relays get
worked on. So, I shouldn't have to do any additional validation after this
rewrite rule executes as my other operations should continue as normal.
My ideal rule would look something like this
*+*@mydomain.ca $1@??? Ff
I'm unclear if this will only look at the sender's field though. I don't
want to apply this rule to a RCPT or To field.
Am I totally out to lunch or should this work?
Thanks,
--
Kevin Vadnais
Systems Programmer
University of Lethbridge
