On 20/12/2022 23:59, Joey J wrote:
Thanks to Bill and Matus for your responses.

Basically, the client is talking about real money transactions, airplanes, paypal etc, but he is a legit sender with these often flagged topics. Sometimes the message goes through, but by the time you reply 2 or 3 times, there are more of the buzz words that SA looks at based on rules.

We can't whitelist j...@company.com because of course everyone pretending to be him will more than likely get whitelisted and you know the rest. This is why I thought if user j...@company.com from ip 1.2.3.4 condition would allow me to add some negative score to get over the total flagging it as spam.

You guys would know better than I as to which would be the best method, I like scoring it some and going to -100.

Within the reject to the user it had the following:

Spam detection results: 3

ClamAVHeuristics 3 ClamAV heuristic test: Phishing.Email.SpoofedDomain (clamav)

AWL -0.969 Adjusted score from AWL reputation of From: address

BAYES_00 -1.9 Bayes spam probability is 0 to 1%

BIGNUM_EMAILS_MANY      2.999 Lots of email addresses/leads, over and over

DKIM_INVALID 0.1 DKIM or DK signature exists, but is not valid

DKIM_SIGNED 0.1 Message has a DKIM or DK signature, not necessarily valid

HTML_FONT_LOW_CONTRAST 0.001 HTML font color similar or identical to background

HTML_MESSAGE 0.001 HTML included in message

KAM_DMARC_STATUS 0.01 Test Rule for DKIM or SPF Failure with Strict Alignment

SPF_HELO_NONE 0.001 SPF: HELO does not publish an SPF Record

SPF_PASS -0.001 SPF: sender matches SPF record

T_FILL_THIS_FORM_SHORT 0.01 Fill in a short form with personal information

URIBL_BLOCKED 0.001 ADMINISTRATOR NOTICE: The query to URIBL was blocked.  See http://wiki.apache.org/spamassassin/DnsBlocklists#dnsbl-block

My approach is like this:

describe LOCAL_WELCOMING_4 Pseudo-welcomelist (case-insensitive)
score LOCAL_WELCOMING_4 -4
header LOCAL_WELCOMING_4 From =~ /(fred\@bloggs\.com|\@jones\.com)>?\s*$/i

I have a few of these with different score reductions (4,6,8,10 etc) all held in /etc/spamassassin/local_welcoming.cf. If you end up with a lot of addresses to be 'welcomed' (as I do) you need some code to manage them, but the principle is simple enough: they act to reduce the score of any email where the 'From:' address matches the regex. They do not guarantee acceptance (the spam score is still calculated, only some amount (4 in the case above) is deducted, and they do not (in my case anyway) apply to virus-laden emails.

Reply via email to