Sorry about that, I used to be much better about not using // and only
/**/, precisely for cc compatibility. I'll go through and fix all the
//'s
For personal installations, the Makefile should be able to read an
environment variable called PREFIX and install there. I think you'll
need to hand-
I'll happily accept patches. In the meantime, killing spamd won't cause
any loss of mail, only loss of identification of spam messages for that
fraction of a second when it's not listening, or for those messages
already in process. spamc will just dump the unprocessed message back
out if process
On Sun, Feb 17, 2002 at 11:22:56AM -0800, Craig Hughes wrote:
| On Sat, 2002-02-16 at 17:21, Erik B. Berry wrote:
| > I've managed to make/compile the 2.01 release under SunOS 5.7,
| > after changing the compiler to gcc from cc (cc doesn't like C++ //
| > comments in C files).
|
| Sorry about that
OK, this isn't something that everyone will be able to use, but it works
nicely for me.
I'm using the courier maildrop LDA. I deliver messages to one recipient at
a time. In maildrop, I've added this:
# Stick the final recipient into the headers
exception {
xfilter "/usr/local/bin/reform
Ok, I did a little bit of searching:
For the envelope FROM, RFC-1123 specifies that the mail server making
"final delivery" of a message:
MUST pass the MAIL FROM: address from the SMTP envelope
with the message, for use if an error notification message must
be
On 17 Feb 2002, Craig Hughes wrote:
> So, for envelope from checking, we should use the "Return-Path" header.
> I'll make a rule which compares Return-Path to From: and see how it does
> at differentiating spam from nonspam.
Hadn't even thought of checking the sender - interesting. I'm curious t
It seems I've been getting a lot of spam lately that has a valid MX, but the
MX is 127.0.0.1 (loopback). Any chance we could add a test for this?
Dan.
___
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/sp
On Sun, Feb 17, 2002 at 03:51:05PM -0700, Charlie Watts wrote:
>
> So ... perhaps we should support both. Envelope-To (which can have
> multiple addresses in it, remember) for folks who can use it, and
> Delivered-To for folks who can't. Simple enough.
>
> The nice thing about doing it with head
How stable is today's (last night's?) CVS?
If I get no complaints, I will release it for Debian - we've needed a
release for a bit.
--
Duncan Findlay
___
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/sp
Should be stable -- no real changes to it for a few days now.
C
On Sun, 2002-02-17 at 17:32, Duncan Findlay wrote:
> How stable is today's (last night's?) CVS?
>
> If I get no complaints, I will release it for Debian - we've needed a
> release for a bit.
>
> --
> Duncan Findlay
>
> _
Has anyone figured out the issue with spamass-filter?
On 17 Feb 2002, Craig Hughes wrote:
> Should be stable -- no real changes to it for a few days now.
>
> C
>
> On Sun, 2002-02-17 at 17:32, Duncan Findlay wrote:
> > How stable is today's (last night's?) CVS?
> >
> > If I get no complaints,
On Wed, Feb 13, 2002 at 08:57:12AM -0800, Craig Hughes wrote:
> Here's the final fix to this problem -- the trick was that the Makefile
> was trying to compile spamc with the same flags that perl itself was
> compiled with, which is overkill (particularly the libs stuff). So I've
> removed the li
On Sun, 17 Feb 2002, Daniel Rogers wrote:
> It seems I've been getting a lot of spam lately that has a valid MX, but
> the MX is 127.0.0.1 (loopback). Any chance we could add a test for
> this?
There was some discussion on the Postfix list a while ago to provide a DNS
Blacklist style lookup for
13 matches
Mail list logo
