Re: [exim] Changing envelope from address to actual recipien…

Inizio della pagina
Delete this message
Reply to this message
Autore: soumya tr
Data:  
To: exim-users@exim.org
Oggetto: Re: [exim] Changing envelope from address to actual recipient address in forwarded mails
Hi Wagner,

Thanks for the response.

Suppose there is a forwarder set say,

mail1@??? -----(forwards all maila to) ----> mail1@???

Now, if say sender@??? tries to send a mail to mail1@???, then it
will be forwarded to mail1@??? (as per the forwarder rules) . In my
exim conf the director which figures out this is:

--------------------------------
virtual_aliases_nostar:
driver = redirect
srs = forward
allow_defer
allow_fail
domains = !$primary_hostname
require_files = "+/etc/valiases/$domain"
address_data = ${lookup{$local_part@
$domain}lsearch{/etc/valiases/$domain}}
data = $address_data
file_transport = address_file
pipe_transport = ${if
forall{/bin/cagefs_enter:/usr/sbin/cagefsctl}{exists{$item}}{cagefs_virtual_address_pipe}{${if
match{${extract{6}{:}{${lookup
passwd{${lookup{$domain}lsearch{/etc/userdomains}{$value}}}{$value}}}}}{\N(jail|no)shell\N}{jailed_virtual_address_pipe}{virtual_address_pipe}}}}
retry_use_local_part
unseen
--------------------------------

All the mails from my server route via a 3rd party server (Mandrillapp). At
mandrill app side they just reject such mails, because the from address for
such mails will be an outside domain (in the above case sender@???).

So what I was trying to do is to set the envelop from header to
mail1@??? (actual header will be sender@???). But that didn't
work.

Another option I was trying is to send all the forwarded mails via server
itself (instead of routing via different server). For this purpose, I tried
to add a header in virtual_aliases_nostar, and added a custom router to
check for the header and send :

--------------------------------

fwded_route:
driver = accept
condition = ${if def:header_X-Fwded-Mail:}
transport = remote_smtp

virtual_aliases_nostar:
driver = redirect
srs = forward
allow_defer
allow_fail
domains = !$primary_hostname
require_files = "+/etc/valiases/$domain"
address_data = ${lookup{$local_part@
$domain}lsearch{/etc/valiases/$domain}}
data = $address_data
condition = ${lookup{$local_part@$domain}lsearch{/etc/valiases/$domain}
{yes}{no}}
headers_add = X-Fwded-mail: ${lookup{$local_part@
$domain}lsearch{/etc/valiases/$domain}}
file_transport = address_file
pipe_transport = ${if
forall{/bin/cagefs_enter:/usr/sbin/cagefsctl}{exists{$item}}{cagefs_virtual_address_pipe}{${if
match{${extract{6}{:}{${lookup
passwd{${lookup{$domain}lsearch{/etc/userdomains}{$value}}}{$value}}}}}{\N(jail|no)shell\N}{jailed_virtual_address_pipe}{virtual_address_pipe}}}}
retry_use_local_part
unseen
--------------------------------

But the above settings doesn't work.

I am not sure if I am going in the right direction. Please assist.



On Wed, May 31, 2017 at 1:42 PM, Ferenc Wágner <wferi@???> wrote:

> soumya tr via Exim-users <exim-users@???> writes:
>
> > On Fri, May 26, 2017 at 12:44 AM, soumya tr <soumya.324@???>
> wrote:
> >
> >> I have a forwarder set. All the outgoing mails relay via a3rd party mail
> >> app . In the 3rd party app I have mentioned the list of allowed domains
> >> that can send out mails.
> >>
> >> But as all the forwarder mails have from address as the actual from, 3rd
> >> party app is rejecting it. So I was thinking of changing the envelop
> header
> >> of the mail to that of actual recipient in the server. Is it possible
> to
> >> achieve it via routers? Is there any other way I can handle this?
> >
> > Has anyone else faced any similar scenario?
>
> I don't know. I tried to understand your description several times, but
> failed. Others may experience a similar difficulty. I suggest you
> rephrase your question using precise terminology and defining ambiguous
> terms like for example "allowed domain", "actual", "envelop header",
> "server" and so on.
> --
> Regards,
> Feri
>




--
Regards,
Soumya
Linux Sytem Administrator
DirectI
<http://poornam.com>
"I like the dreams of the future better than the history of the past."