https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6089
Summary: Problem with RATWARE_EFROM test
Product: Spamassassin
Version: unspecified
Platform: Other
OS/Version: All
Status: NEW
Severity: enhancement
Priority: P5
Component: Rules
AssignedTo: [email protected]
ReportedBy: [email protected]
My own outgoing mails are being marked with RATWARE_EFROM, because I use a form
of SRS on them. Instead of sending MAIL FROM:<[email protected]>, I always
use a reverse-path of the form
<srs0+cryptohash+timestamp+infradead.org+dw...@hostname.srs.infradead.org>.
This means that I don't accept bounces to [email protected], since they'd all
be in response to fake messages. And it means that anyone doing sender
verification callouts will reject fake mail from [email protected] too.
This rewriting scheme is taken from the early development of 'SRS' -- the
Sender Rewriting Scheme which the SPF advocates imagined all mail forwarders
would implement in order to confirm with their new world order.
My understanding of perl is limited, but you seem to be trying to compare the
(envelope) sender and (header) recipient addresses somehow -- as if I'm only
allowed to send email to myself, perhaps?
Is there a coherent explanation somewhere of what this test is trying to do?
It seems completely bogus to me. The address encoded in the envelope sender is
the original _sender_ address, and bears no relation to the recipient in the
To: header, who may not be an actual recipient of the mail at all (it could
Resent-To: someone else). So comparing the two seems... odd.
Even if you were comparing with the From: (or Resent-From:) address, its not a
good test -- for all the same reasons that you don't check that envelope sender
== header sender.
--
Configure bugmail:
https://issues.apache.org/SpamAssassin/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
