Pure WorldWide Internet
510 W. RiverSide Ave
Suite 208
Spokane, Wa. 99201
IncrediMail - Email has finally evolved - Click Here
as you point out, the problem is spammers can forge what's in the helo
message just as they forge what's in MAIL FROM.
but also, unfortunately, a way large percentage of sites do not have
correctly configured names in their helos.
(some have ip addresses. some have their non-fully-qualified name
does anyone else have a need for finer-grained control over rbl
lookups?
i have a complicated dns setup in one installation that runs sa
on around 100k messages/day.
it's a typical split dns setup with one name server for the publicly
visible names (running bind 9), and another (microsoft on one
most people who actually have done usability testing discover that
almost nobody *reads* documentation (even readme files), though some
people attempt to refer to them to answer questions. they instead
prefer to use well-developed intuitions. what happens when the
software is counterintuitive is
sure he can complain to his isp, but...
i am amazed that, in this day and age, software developers will not
listen to usability complaints for what they are (referring to the
several replies to this note already posted on this list).
it would suggest, or you could infer, that the default install
speaking of which, are there isp-specific lists which distinguish between the isp's
customer netblocks and the isp's internal netblocks?
(the objective being to block smtp-talking clients and force them to relay through
their isp's smtp relays.)
On Wed, Mar 19, 2003 at 12:22:26PM -0800, Christian
On Wed, Jan 15, 2003 at 05:13:26PM -0600, Jeremy Turner wrote:
> Easy enough. I have my settings to not clobber HTML emails, and put
> reports in the header. I use an MTA-level subject rewrite so that
> emails within a certain spam range get classified as 'Yellow' or 'Red'.
> Therefore, a spam
On Wed, Jan 15, 2003 at 10:42:07PM -, Stephane wrote:
> Hello again,
>
> Amongst the answers to my previous post ("has a large company implemented SA") there
>was a very good idea on success stories with SA... It is true that most of the people
>on the net who would say they were happy with
thank you and martin for quantitating this.
but...
should i read this as 76.8% of total messages received were caught by
razor?
(should i believe that only 100%-97.4% of your messages are non-spam,
only 111 messages in the last month? maybe i don't understand what the
*total* represents.)
(als
i gather you are looking to reduce the cost of spam filtering,
by cutting out inefficient or costly tests.
my impression (not from measuring, just from looking at spam) is that
the checksum databases are relatively inefficient because spammers
have started to routinely introduce random components
ad of writing your own?
On Mon, Nov 04, 2002 at 11:42:01AM -0800, Harlan Limajliew - MIS wrote:
> The following is strictly to represent the scenario -
>
>
> Goal:
>
> Replacing our Windows based filter (if you could call it that) that is using
> a program named Mail Essent
The following is strictly to represent the scenario -
Goal:
Replacing our Windows based filter (if you could call it
that) that is using a program named Mail Essentials. It does not provide anywhere near the
throughput we need (~120/min) and I have been hoping to show our IT di
autowhitelist is not supported by amavisd.
when i used autowhitelist out of pracmail on solaris, it would hang
fairly regularly with a few hundred processes queued up on the database
being locked.
i like the idea of autowhitelist for some applications, for example,
individual users with email ad
what do people think of the concept of a graylist?
i am managing the spam filtering for a sizeable enterprise of
knowledge workers many of whom bill by the hour. for such workers
there's a good time-management argument for email filtering.
i'm trying to avoid supporting individual preferences a
this problem also occurs if you have your mail forwarded from other
sites that run sa.
my personal solution is to put the hostname of the sa reporter in the
report header, so i can recognize where the report is coming from,
and then choose to accept or delete it.
(i add it as a comment within
i agree with you, particularly now that i've counted that (off hours)
2/3 of the mail being delivered to a company whose firewalls i manage
was graded as spam by sa. at the point we needed to add an extra
machine purely for spam assassin, i decided to get serious about blocking,
which i'd previou
i have observed numbers in the range of 50% spam for several
enterprises measured over the day.
i use amavisd and postfix for this purpose.
i set loglevel=3 in /etc/amavisd.conf.
then here's a shell script that computes it:
#!/bin/sh
log=/s3/amavis/amavis.log
spamin=`grep $1 $log | grep "spam
17 matches
Mail list logo
