Re: [exim-dev] [Bug 2702] New: XCLIENT ESMTP extension

Pàgina inicial
Delete this message
Reply to this message
Autor: Vsevolod Stakhov
Data:  
A: exim-dev
Assumpte: Re: [exim-dev] [Bug 2702] New: XCLIENT ESMTP extension
On 22/02/2021 14:23, admin--- via Exim-dev wrote:
> https://bugs.exim.org/show_bug.cgi?id=2702
>
>             Bug ID: 2702
>            Summary: XCLIENT ESMTP extension
>            Product: Exim
>            Version: 4.94
>           Hardware: x86
>                 OS: Linux
>             Status: NEW
>           Severity: wishlist
>           Priority: low
>          Component: Mail Receipt
>           Assignee: unallocated@???
>           Reporter: jgh146exb@???
>                 CC: exim-dev@???

>
> A fork of the Exim project is carrying (among others) a patch adding XCLIENT
> support, per http://www.postfix.org/XCLIENT_README.html . We might consider
> doing
> the same. This is an inbound proxy method, like Proxy Protocol but triggered
> by
> an SMTP command.
>
> The patch
>
> https://github.com/SpamExperts/exim/commit/3798d48d73c89f7835726d31f096851f7f7fca2a
> isn't immediately usable:
>
> - Proper handling of the proxy address/port details, for logging
> - We should consider re-calling the connect ACL, after deciding to accept the
> XCLIENT command, to give the chance to re-evaluate connect-time decisions
> with the proxy-supplied info for the connection
> - Ditto re-call the helo ACL, with the HELO attribute value
> - We could consider a dedicated ACL for the command, separate from the
> allowed-hosts list (still needed for advertise)
> - It should be a compile-time option, initially Experimental
> - duplicated code: xclient_xtextdecode() vs. auth_xtextdecode()
> - (nit) coding style
>


(Resending from the correct email for this ML)

Is it anyhow different from a patch I have proposed some (long) time
ago?
https://mta.org.ua/exim-4.76-conf/packages/ports-freebsd/exim-4.83_2/core/files/extra-patch-xclient

I have stopped to support it because the Exim developers were too
reluctant to include that into the main distribution, so I have also
removed it from the FreeBSD port. The code looks quite familiar from
what I see...