IBM AH0QXML User Guide - Page 108

96

Lotus Domino 6 spam Survival Guide for IBM eServer

rating. Positive tests identify characteristics of legitimate messages, while

negative tests identify characteristics of spam messages. When combined, they

yield extremely efficient spam identification. Following are examples of the types

of tests conducted to identify the probability that a message is spam:

±

Basic keywords and phrases in the e-mail message, like AMAZING or casino

±

Malformed headers or addresses in the e-mail message header, such as

missing headers, invalid date, suspicious list of recipients, recipients

commonality ratio, forged

“

Received

”

headers

±

Pornography rules - approximately 40 positive and negative tests to prevent

false positives

±

HTML form use in e-mail body

±

Usual spam claims and unsubscribe URLs, such as

“

to be removed..

”

,

“

we do

not send...

”

,

“

http://...?remove=...

”

±

Complex text patterns for common spam content, such as (?:You (?:were

sent|have received|are receiving)|You're

receiving).{0,15}(?:message|e-?mail)s? because - if you

(?:(?:want|wish|care|prefer) not to |(?:don't|do not) (?:want|wish|care) to

)(?:be contacted again|receive (any)?\s*(?:more|future|further)

(?:e?-?mail|messages?|offers|solicitations))

±

Foreign character sets detected in headers and body

±

Spam phrase identification. Count the ratio of frequent spam phrase

occurrences, such as

“

credit card,

”

“

loan you,

”

“

multi level

”

and compare it with

threshold value

±

Unique identifiers, tags in the message/subject

±

Suspicious text formats, for example a gap in text ("G E T

R I C H

F A S T"),

lines of yelling ( BUY NOW!!! )

±

Nigerian scam, multiple patterns

The Nigerian scam always mutates, and can be detected by pattern matching

only.

±

Network checks such as no MX record for sender's domain

±

Realtime Blacklist checks, 10 RBL systems supported

RBL tests don't uniquely identify a message as spam. They act as a

contributing test to the probability that a given message is spam.

±

Distributed checksum checks

Razor network check supported. Provides peer-based identification of spam

messages.

Section	Page
Front cover	1
Contents	5
Notices	7
Trademarks	8
Preface	9
The team that wrote this redbook	9
Become a published author	11
Comments welcome	12
Chapter 1. Introduction	13
1.1 Definition of spam	14
1.2 Categorizing spam	14
Chapter 2. Preventing unwanted e-mail and spam	19
2.1 Spam avoidance techniques	20
2.1.1 Passive harvesting attacks	20
2.1.2 Avoiding harvesting	21
2.1.3 Confusing the harvesters	22
2.1.4 Inform Users	22
2.2 How to block spam	24
2.2.1 At the gateway	25
2.2.2 At the server	25
2.2.3 By the end user	26
2.2.4 Selecting the best approach	26
2.2.5 Managing the ongoing anti-spam campaign	27
2.2.6 Summary	28
Chapter 3. Domino 6 anti-spam architecture	29
3.1 The Domino messaging environment	30
3.2 Domino 6 messaging components	32
3.2.1 SMTP Listener/Server	33
3.2.2 Router	35
3.3 Domino 6 anti-spam configuration	35
3.3.1 Server configuration features	36
3.3.2 User configuration features	37
3.4 Common problems and solutions	37
Chapter 4. Domino 6 Server anti-spam features	41
4.1 How to detect spam	42
4.1.1 Examining the message properties	42
4.1.2 Separating legitimate e-mail from spam based on content	43
4.2 Controlling connections from spammers	44
4.2.1 DNS Blacklist filters	44
4.2.2 Inbound Intended Recipients Controls	48
4.2.3 Disable routing mail to groups	51
4.2.4 Inbound connection controls	53
4.2.5 Inbound sender controls	54
4.3 Controlling delivery of spam	57
4.3.1 Server mail rules	57
4.4 Controlling use of your server as a relay	66
4.4.1 Inbound relay controls	66
4.4.2 Inbound relay enforcement	71
4.5 Protecting your Domino server from active address harvesting attacks	74
4.5.1 SMTP harvesting attacks	74
4.5.2 Spam mail bombing	74
4.5.3 Direct SMTP RCPT TO harvesting	74
4.5.4 Defending against active attacks	75
Chapter 5. Using mail file rules to prevent spam	77
5.1 Distinguishing between spam and legitimate e-mail	78
5.2 Mail file rules	80
5.2.1 Setting up mail file rules	80
5.2.2 Developing anti-spam mail file rules	83
5.2.3 Viewing mail rules and the evaluation sequence	89
5.2.4 Monitoring mail file rules	90
Chapter 6. Third party anti-spam products	93
6.1 Anti-spam products for Notes and Domino	94
6.1.1 spamJam for Lotus Notes and Domino	94
6.1.2 SpamEraser for Lotus Notes and Domino 6	97
6.1.3 iQ.Suite	100
6.1.4 ScanMail for Lotus Notes with eManager	101
6.1.5 XM SpamStop	104
6.1.6 Other anti-spam products for Notes and Domino	106
6.2 Anti-spam server and gateway products and services	106
6.2.1 BrightMail Anti-Spam 4.0 from BrightMail, Inc.	106
6.2.2 ActiveState PureMessage	107
6.2.3 Trend Micro Inc. products	109
6.2.4 Other anti-spam server or gateway products and services	110
Related publications	111
IBM Redbooks	111
Referenced Web sites	111
How to get IBM Redbooks	112
IBM Redbooks collections	112
Index	113

IBM AH0QXML User Guide - Page 108

and compare it with, threshold value

Page 108 highlights