Mercurial > dnsbl
annotate xml/dnsbl.in @ 201:752d4315675c stable-6-0-16
add reference to mercurial repository in the documentation
| author | Carl Byington <carl@five-ten-sg.com> |
|---|---|
| date | Sat, 02 Feb 2008 12:31:53 -0800 |
| parents | f0eda59e8afd |
| children | 92a5c866bdfa |
| rev | line source |
|---|---|
| 108 | 1 <reference> |
| 2 <title>@PACKAGE@ Sendmail milter - Version @VERSION@</title> | |
| 3 <partintro> | |
| 4 <title>Packages</title> | |
|
201
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
5 |
| 108 | 6 <para>The various source and binary packages are available at <ulink |
|
201
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
7 url="http://www.five-ten-sg.com/@PACKAGE@/packages/">http://www.five-ten-sg.com/@PACKAGE@/packages/</ulink>. |
| 108 | 8 The most recent documentation is available at <ulink |
|
201
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
9 url="http://www.five-ten-sg.com/@PACKAGE@/">http://www.five-ten-sg.com/@PACKAGE@/</ulink>. |
|
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
10 </para> |
|
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
11 |
|
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
12 <para>A <ulink |
|
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
13 url="http://www.selenic.com/mercurial/wiki/">Mercurial</ulink> source |
|
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
14 code repository for this project is available at <ulink |
|
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
15 url="http://hg.five-ten-sg.com/@PACKAGE@/">http://hg.five-ten-sg.com/@PACKAGE@/</ulink>. |
| 108 | 16 </para> |
| 94 | 17 |
| 108 | 18 </partintro> |
| 94 | 19 |
| 108 | 20 <refentry id="@PACKAGE@.1"> |
| 21 <refentryinfo> | |
|
201
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
22 <date>2008-02-02</date> |
| 108 | 23 </refentryinfo> |
| 94 | 24 |
| 108 | 25 <refmeta> |
| 26 <refentrytitle>@PACKAGE@</refentrytitle> | |
| 27 <manvolnum>1</manvolnum> | |
| 28 <refmiscinfo>@PACKAGE@ @VERSION@</refmiscinfo> | |
| 29 </refmeta> | |
| 30 | |
| 31 <refnamediv id='name.1'> | |
| 32 <refname>@PACKAGE@</refname> | |
| 33 <refpurpose>a sendmail milter with per-user dnsbl filtering</refpurpose> | |
| 34 </refnamediv> | |
| 94 | 35 |
| 108 | 36 <refsynopsisdiv id='synopsis.1'> |
| 37 <title>Synopsis</title> | |
| 38 <cmdsynopsis> | |
| 39 <command>@PACKAGE@</command> | |
| 40 <arg><option>-c</option></arg> | |
| 41 <arg><option>-s</option></arg> | |
| 42 <arg><option>-d <replaceable class="parameter">n</replaceable></option></arg> | |
| 43 <arg><option>-e <replaceable class="parameter">from|to</replaceable></option></arg> | |
| 179 | 44 <arg><option>-b <replaceable class="parameter">local-domain-socket</replaceable></option></arg> |
| 108 | 45 <arg><option>-r <replaceable class="parameter">local-domain-socket</replaceable></option></arg> |
| 46 <arg><option>-p <replaceable class="parameter">sendmail-socket</replaceable></option></arg> | |
| 47 <arg><option>-t <replaceable class="parameter">timeout</replaceable></option></arg> | |
| 48 </cmdsynopsis> | |
| 49 </refsynopsisdiv> | |
| 94 | 50 |
| 108 | 51 <refsect1 id='options.1'> |
| 52 <title>Options</title> | |
| 53 <variablelist> | |
| 54 <varlistentry> | |
| 55 <term>-c</term> | |
| 111 | 56 <listitem><para> |
| 57 Load the configuration file, print a cannonical form | |
| 58 of the configuration on stdout, and exit. | |
| 59 </para></listitem> | |
| 108 | 60 </varlistentry> |
| 61 <varlistentry> | |
| 62 <term>-s</term> | |
| 111 | 63 <listitem><para> |
| 64 Stress test the configuration loading code by repeating | |
| 65 the load/free cycle in an infinite loop. | |
| 66 </para></listitem> | |
| 108 | 67 </varlistentry> |
| 68 <varlistentry> | |
| 69 <term>-d <replaceable class="parameter">n</replaceable></term> | |
| 111 | 70 <listitem><para> |
| 71 Set the debug level to <replaceable class="parameter">n</replaceable>. | |
| 72 </para></listitem> | |
| 108 | 73 </varlistentry> |
| 74 <varlistentry> | |
| 75 <term>-e <replaceable class="parameter">from|to</replaceable></term> | |
| 111 | 76 <listitem><para> |
| 77 Print the results of looking up the from and to addresses in the | |
| 78 current configuration. The | character is used to separate the from and to | |
| 79 addresses in the argument to the -e switch. | |
| 80 </para></listitem> | |
| 108 | 81 </varlistentry> |
| 82 <varlistentry> | |
| 179 | 83 <term>-b <replaceable class="parameter">local-domain-socket-file-name</replaceable></term> |
| 84 <listitem><para> | |
| 85 Set the local socket used for the connection to the dccifd daemon. | |
| 86 This is typically /var/dcc/dccifd. | |
| 87 </para></listitem> | |
| 88 </varlistentry> | |
| 89 <varlistentry> | |
| 90 <term>-r <replaceable class="parameter">local-domain-socket-file-name</replaceable></term> | |
| 111 | 91 <listitem><para> |
| 92 Set the local socket used for the connection to our own dns resolver processes. | |
| 93 </para></listitem> | |
| 108 | 94 </varlistentry> |
| 95 <varlistentry> | |
| 96 <term>-p <replaceable class="parameter">sendmail-socket</replaceable></term> | |
| 111 | 97 <listitem><para> |
| 98 Set the socket used for the milter connection to sendmail. This is either | |
| 99 "inet:port@ip-address" or "local:local-domain-socket-file-name". | |
| 100 </para></listitem> | |
| 108 | 101 </varlistentry> |
| 102 <varlistentry> | |
| 103 <term>-t <replaceable class="parameter">timeout</replaceable></term> | |
| 111 | 104 <listitem><para> |
| 105 Set the timeout in seconds used for communication with sendmail. | |
| 106 </para></listitem> | |
| 108 | 107 </varlistentry> |
| 108 </variablelist> | |
| 109 </refsect1> | |
| 94 | 110 |
| 111 | 111 <refsect1 id='usage.1'> |
| 108 | 112 <title>Usage</title> |
| 113 <para><command>@PACKAGE@</command> -c</para> | |
| 114 <para><command>@PACKAGE@</command> -s</para> | |
| 111 | 115 <para><command>@PACKAGE@</command> -e 'someone@aol.com|localname@mydomain.tld'</para> |
| 116 <para><command>@PACKAGE@</command> -d 10 -r resolver.sock -p local:dnsbl.sock</para> | |
| 117 </refsect1> | |
| 118 | |
| 119 <refsect1 id='installation.1'> | |
| 120 <title>Installation</title> | |
| 121 <para> | |
| 122 This is now a standard GNU autoconf/automake installation, so the normal | |
| 123 "./configure; make; su; make install" works. "make chkconfig" will | |
| 124 setup the init.d runlevel scripts. Alternatively, you can use the | |
| 125 source or binary RPMs at <ulink | |
| 126 url="http://www.five-ten-sg.com/@PACKAGE@/packages">http://www.five-ten-sg.com/@PACKAGE@/packages</ulink>. | |
| 127 </para> | |
| 128 <para> | |
| 129 Note that this has ONLY been tested on Linux, specifically RedHat Linux. | |
| 130 In particular, this milter makes no attempt to understand IPv6. Your | |
| 131 mileage will vary. You will need at a minimum a C++ compiler with a | |
| 132 minimally thread safe STL implementation. The distribution includes a | |
| 133 test.cpp program. If it fails this milter won't work. If it passes, | |
| 134 this milter might work. | |
| 135 </para> | |
| 136 <para> | |
| 137 Modify your sendmail.mc by removing all the "FEATURE(dnsbl" lines, add | |
| 138 the following line in your sendmail.mc and rebuild the .cf file | |
| 139 </para> | |
| 140 <para><screen>INPUT_MAIL_FILTER(`dnsbl', `S=local:/var/run/dnsbl/dnsbl.sock, F=T, T=C:30s;S:5m;R:5m;E:5m')</screen></para> | |
| 141 <para> | |
| 142 Modify the default <citerefentry> | |
| 143 <refentrytitle>@PACKAGE@.conf</refentrytitle> <manvolnum>5</manvolnum> | |
| 144 </citerefentry> configuration. | |
| 145 </para> | |
| 146 </refsect1> | |
| 147 | |
| 148 <refsect1 id='configuration.1'> | |
| 149 <title>Configuration</title> | |
| 150 <para> | |
| 151 The configuration file is documented in <citerefentry> | |
| 152 <refentrytitle>@PACKAGE@.conf</refentrytitle> <manvolnum>5</manvolnum> | |
| 153 </citerefentry>. Any change to the config file, or any file included | |
| 154 from that config file, will cause it to be reloaded within three | |
| 155 minutes. | |
| 156 </para> | |
| 108 | 157 </refsect1> |
| 94 | 158 |
| 108 | 159 <refsect1 id='introduction.1'> |
| 160 <title>Introduction</title> | |
| 161 <para> | |
| 162 Consider the case of a mail server that is acting as secondary MX for a | |
| 163 collection of clients, each of which has a collection of mail domains. | |
| 164 Each client may use their own collection of DNSBLs on their primary mail | |
| 165 server. We present here a mechanism whereby the backup mail server can | |
| 166 use the correct set of DNSBLs for each recipient for each message. As a | |
| 167 side-effect, it gives us the ability to customize the set of DNSBLs on a | |
| 183 | 168 per-recipient basis, so that fred@example.com could use LOCAL and the |
| 108 | 169 SBL, where all other users @example.com use only the SBL. |
| 170 </para> | |
| 171 <para> | |
| 172 This milter can also verify the envelope from/recipient pairs with the | |
| 173 primary MX server. This allows the backup mail servers to properly | |
| 174 reject mail sent to invalid addresses. Otherwise, the backup mail | |
| 175 servers will accept that mail, and then generate a bounce message when | |
| 176 the message is forwarded to the primary server (and rejected there with | |
| 127 | 177 no such user). These rejections are the primary cause of such backscatter. |
| 108 | 178 </para> |
| 179 <para> | |
| 180 This milter will also decode (uuencode, base64, mime, html entity, url | |
| 181 encodings) and scan for HTTP and HTTPS URLs and bare hostnames in the | |
| 182 body of the mail. If any of those host names have A or NS records on | |
| 183 the SBL (or a single configurable DNSBL), the mail will be rejected | |
| 184 unless previously whitelisted. This milter also counts the number of | |
| 185 invalid HTML tags, and can reject mail if that count exceeds your | |
| 186 specified limit. | |
| 187 </para> | |
| 188 <para> | |
| 136 | 189 This milter can also impose hourly rate limits on the number of |
| 190 recipients accepted from SMTP AUTH connections, that would otherwise be | |
| 191 allowed to relay thru this mail server with no spam filtering. | |
| 192 </para> | |
| 193 <para> | |
| 162 | 194 Consider the case of a message from A to B passing thru this milter. If |
| 195 that message is not blocked, then we might eventually see a reply | |
| 156 | 196 message from B to A. If the filtering context for A includes an |
| 162 | 197 autowhite entry, and that context does <emphasis>not</emphasis> cover B |
| 198 as a recipient, then this milter will add an entry in that file to | |
| 199 whitelist such replies for a configurable time period. Suppose A and B | |
| 200 are in the same domain, or at least use the same filtering context. In | |
| 201 that case we don't want to add a whitelist entry for B, since that would | |
| 202 then allow spammers to send mail from B (forged) to B. Such autowhite | |
| 160 | 203 files need to be writeable by the dnsbl user, where all the other dnsbl |
| 204 configuration files only need to be readable by the dnsbl user. | |
| 156 | 205 </para> |
| 206 <para> | |
|
176
4ec928b24bab
allow manual whitelisting with stamp 1 to remove a whitelist entry
carl
parents:
175
diff
changeset
|
207 You can manually add such an autowhite entry, by appending a single |
|
4ec928b24bab
allow manual whitelisting with stamp 1 to remove a whitelist entry
carl
parents:
175
diff
changeset
|
208 text line to the autowhitelist file, using something like |
|
4ec928b24bab
allow manual whitelisting with stamp 1 to remove a whitelist entry
carl
parents:
175
diff
changeset
|
209 <command>echo "$mail 0" >>$autowhitefile</command>. |
|
4ec928b24bab
allow manual whitelisting with stamp 1 to remove a whitelist entry
carl
parents:
175
diff
changeset
|
210 You can manually remove such an autowhite entry, by appending a single |
|
4ec928b24bab
allow manual whitelisting with stamp 1 to remove a whitelist entry
carl
parents:
175
diff
changeset
|
211 text line to the autowhitelist file, using something like |
|
4ec928b24bab
allow manual whitelisting with stamp 1 to remove a whitelist entry
carl
parents:
175
diff
changeset
|
212 <command>echo "$mail 1" >>$autowhitefile</command>. |
|
4ec928b24bab
allow manual whitelisting with stamp 1 to remove a whitelist entry
carl
parents:
175
diff
changeset
|
213 </para> |
|
4ec928b24bab
allow manual whitelisting with stamp 1 to remove a whitelist entry
carl
parents:
175
diff
changeset
|
214 <para> |
| 108 | 215 The DNSBL milter reads a text configuration file (dnsbl.conf) on |
| 216 startup, and whenever the config file (or any of the referenced include | |
| 217 files) is changed. The entire configuration file is case insensitive. | |
| 218 If the configuration cannot be loaded due to a syntax error, the milter | |
| 219 will log the error and quit. If the configuration cannot be reloaded | |
| 220 after being modified, the milter will log the error and send an email to | |
| 152 | 221 root from dnsbl@$hostname. You probably want to add dnsbl@$hostname |
| 108 | 222 to your /etc/mail/virtusertable since otherwise sendmail will reject |
| 223 that message. | |
| 224 </para> | |
| 225 </refsect1> | |
| 94 | 226 |
| 111 | 227 <refsect1 id='dcc.1'> |
| 108 | 228 <title>DCC Issues</title> |
| 229 <para> | |
| 230 If you are also using the <ulink | |
| 231 url="http://www.rhyolite.com/anti-spam/dcc/">DCC</ulink> milter, there | |
| 232 are a few considerations. You may need to whitelist senders from the | |
| 233 DCC bulk detector, or from the DNS based lists. Those are two very | |
| 234 different reasons for whitelisting. The former is done thru the DCC | |
| 235 whiteclnt config file, the later is done thru the DNSBL milter config | |
| 236 file. | |
| 237 </para> | |
| 238 <para> | |
| 239 You may want to blacklist some specific senders or sending domains. | |
| 240 This could be done thru either the DCC (on a global basis, or for a | |
| 241 specific single recipient). We prefer to do such blacklisting via the | |
| 242 DNSBL milter config, since it can be done for a collection of recipient | |
| 243 mail domains. The DCC approach has the feature that you can capture the | |
| 244 entire message in the DCC log files. The DNSBL milter approach has the | |
| 245 feature that the mail is rejected earlier (at RCPT TO time), and the | |
| 246 sending machine just gets a generic "550 5.7.1 no such user" message. | |
| 247 </para> | |
| 248 <para> | |
| 249 The DCC whiteclnt file can be included in the DNSBL milter config by the | |
| 250 dcc_to and dcc_from statements. This will import the (env_to, env_from, | |
| 251 and substitute mail_host) entries from the DCC config into the DNSBL | |
| 252 config. This allows using the DCC config as the single point for | |
| 253 white/blacklisting. | |
| 254 </para> | |
| 255 <para> | |
| 256 Consider the case where you have multiple clients, each with their own | |
| 257 mail servers, and each running their own DCC milters. Each client is | |
| 258 using the DCC facilities for envelope from/to white/blacklisting. | |
| 259 Presumably you can use rsync or scp to fetch copies of your clients DCC | |
| 260 whiteclnt files on a regular basis. Your mail server, acting as a | |
| 261 backup MX for your clients, can use the DNSBL milter, and include those | |
| 262 client DCC config files. The envelope from/to white/blacklisting will | |
| 263 be appropriately tagged and used only for the domains controlled by each | |
| 264 of those clients. | |
| 265 </para> | |
| 179 | 266 <para> |
| 267 You can now use (via dccifd) different dcc filtering parameters on a per | |
| 268 context basis. See the dcc_greylist and dcc_bulk_threshold statements | |
| 269 in the <citerefentry> <refentrytitle>@PACKAGE@.conf</refentrytitle> | |
| 270 <manvolnum>5</manvolnum> </citerefentry> configuration. Those | |
| 271 statements are only active if you supply the <option>-b</option> option | |
| 272 on the dnsbl command line. If you use the dcc via the standard dcc | |
| 273 milter (dccm), then connections from clients that use SMTP AUTH are | |
| 274 still subject to greylisting. If you use the dcc via dccifd and this | |
| 275 milter, then connections from clients that use SMTP AUTH are never | |
| 180 | 276 subject to greylisting. As part of this per-user greylisting, you need |
| 277 to move the dnsblnogrey file from the config directory to something | |
| 278 like /var/dcc/userdirs/local/dnsblnogrey/whiteclnt so the dccifd will | |
| 279 properly ignore greylisting for those recipients that don't want it. | |
| 179 | 280 </para> |
| 108 | 281 </refsect1> |
| 94 | 282 |
| 111 | 283 <refsect1 id='definitions.1'> |
| 108 | 284 <title>Definitions</title> |
| 285 <para> | |
| 286 CONTEXT - a collection of parameters that defines the filtering context | |
| 287 to be used for a collection of envelope recipient addresses. The | |
| 288 context includes such things as the list of DNSBLs to be used, and the | |
| 289 various content filtering parameters. | |
| 290 </para> | |
| 291 <para> | |
| 292 DNSBL - a named DNS based blocking list is defined by a dns suffix (e.g. | |
| 293 sbl-xbl.spamhaus.org) and a message string that is used to generate the | |
| 294 "550 5.7.1" smtp error return code. The names of these DNSBLs will be | |
| 295 used to define the DNSBL-LISTs. | |
| 296 </para> | |
| 297 <para> | |
| 298 DNSBL-LIST - a named list of DNSBLs that will be used for specific | |
| 299 recipients or recipient domains. | |
| 300 </para> | |
| 301 </refsect1> | |
| 94 | 302 |
| 111 | 303 <refsect1 id='filtering.1'> |
| 108 | 304 <title>Filtering Procedure</title> |
| 305 <para> | |
| 152 | 306 The SMTP envelope 'from' and 'to' values are used in various checks. |
| 307 The first check is to see if a reply message (swapping the env_from and | |
| 160 | 308 env_to values) would be unconditionally blocked (just based on the |
| 309 envelope from address). That check is similar to the main check | |
| 310 described below, but there is no body content to be scanned, and there | |
| 311 is no client connection ip address to be checked against DNSBLs. If | |
| 312 such a reply message would be blocked, we also block the original | |
| 313 outgoing message. This prevents folks from sending mail to recipients | |
| 314 that are unable to reply. | |
| 152 | 315 </para> |
| 316 <para> | |
| 136 | 317 If the client has authenticated with sendmail, the rate limits are |
| 318 checked. If the authenticated user has not exceeded the hourly rate | |
|
144
31ff00ea6bfb
allow parent/child to share a fully qualified env_to address
carl
parents:
140
diff
changeset
|
319 limit, then the mail is accepted, the filtering contexts are not used, |
| 136 | 320 the dns lists are not checked, and the body content is not scanned. If |
| 321 the client has not authenticated with sendmail, we follow these steps | |
| 322 for each recipient. | |
| 108 | 323 </para> |
| 324 <orderedlist> | |
| 111 | 325 <listitem><para> |
| 108 | 326 The envelope to email address is used to find an initial filtering |
| 327 context. We first look for a context that specified the full email | |
| 328 address in the env_to statement. If that is not found, we look for a | |
| 329 context that specified the entire domain name of the envelope recipient | |
| 330 in the env_to statement. If that is not found, we look for a context | |
| 331 that specified the user@ part of the envelope recipient in the env_to | |
| 332 statement. If that is not found, we use the first top level context | |
| 333 defined in the config file. | |
| 111 | 334 </para></listitem> |
| 335 <listitem><para> | |
| 108 | 336 The initial filtering context may redirect to a child context based on |
| 337 the values in the initial context's env_from statement. We look for [1) | |
| 338 the full envelope from email address, 2) the domain name part of the | |
| 339 envelope from address, 3) the user@ part of the envelope from address] | |
| 340 in that context's env_from statement, with values that point to a child | |
| 341 context. If such an entry is found, we switch to that child filtering | |
| 342 context. | |
| 111 | 343 </para></listitem> |
| 344 <listitem><para> | |
| 108 | 345 We lookup [1) the full envelope from email address, 2) the domain name |
| 346 part of the envelope from address, 3) the user@ part of the envelope | |
| 347 from address] in the filtering context env_from statement. That results | |
| 348 in one of (white, black, unknown, inherit). | |
| 111 | 349 </para></listitem> |
| 350 <listitem><para> | |
| 108 | 351 If the answer is black, mail to this recipient is rejected with "no such |
| 352 user", and the dns lists are not checked. | |
| 111 | 353 </para></listitem> |
| 354 <listitem><para> | |
| 108 | 355 If the answer is white, mail to this recipient is accepted and the dns |
| 356 lists are not checked. | |
| 111 | 357 </para></listitem> |
| 358 <listitem><para> | |
| 108 | 359 If the answer is unknown, we don't reject yet, but the dns lists will be |
| 360 checked, and the content may be scanned. | |
| 111 | 361 </para></listitem> |
| 362 <listitem><para> | |
| 108 | 363 If the answer is inherit, we repeat the envelope from search in the |
| 364 parent context. | |
| 111 | 365 </para></listitem> |
| 366 <listitem><para> | |
| 168 | 367 If the mail has not been accepted or rejected yet, the dns lists |
| 368 specified in the filtering context are checked and the mail is rejected | |
| 369 if any list has an A record for the standard dns based lookup scheme | |
| 370 (reversed octets of the client followed by the dns suffix). | |
| 371 </para></listitem> | |
| 372 <listitem><para> | |
| 373 If the mail has not been accepted or rejected yet, and the filtering | |
| 170 | 374 context (or any ancestor context) specifies a non-empty generic regular |
| 375 expression, then we check the fully qualified client name (obtained via | |
| 376 the sendmail macro "_"). The mail is rejected if the client name | |
| 377 matches the specified regular expression. | |
| 111 | 378 </para></listitem> |
| 379 <listitem><para> | |
| 108 | 380 If the mail has not been accepted or rejected yet, we look for a |
| 381 verification context, which is the closest ancestor of the filtering | |
| 382 context that both specifies a verification host, and which covers the | |
| 383 envelope to address. If we find such a verification context, and the | |
| 384 verification host is not our own hostname, we open an smtp conversation | |
| 385 with that verification host. The current envelope from and recipient to | |
| 386 values are passed to that verification host. If we receive a 5xy | |
| 387 response those commands, we reject the current recipient with "no such | |
| 388 user". | |
| 111 | 389 </para></listitem> |
| 390 <listitem><para> | |
| 108 | 391 If the mail has not been accepted or rejected yet, and the filtering |
| 392 context enables content filtering, and this is the first such recipient | |
| 393 in this smtp transaction, we set the content filtering parameters from | |
| 394 this context, and enable content filtering for the body of this message. | |
| 111 | 395 </para></listitem> |
| 108 | 396 </orderedlist> |
| 397 <para> | |
| 160 | 398 For each recipient that was accepted, we search for an autowhite entry |
| 399 starting in the reply filtering context. If an autowhite entry is found, | |
| 400 we add the recipient to that auto whitelist file. This will prevent reply | |
| 401 messages from being blocked by the dnsbl or content filtering. | |
| 402 </para> | |
| 403 <para> | |
| 108 | 404 If content filtering is enabled for this body, the mail text is decoded |
| 119 | 405 (uuencode, base64, mime, html entity, url encodings), and scanned for HTTP |
| 406 and HTTPS URLs or bare host names. Hostnames must be either ip address | |
| 407 literals, or must end in a string defined by the TLD list. The first | |
| 408 <configurable> host names are checked as follows. | |
| 409 </para> | |
| 410 <para> | |
| 411 The only known list that is suitable for the content filter DNSBL is the | |
| 412 SBL. If the content filter DNSBL is defined, and any of those host | |
| 413 names resolve to ip addresses that are on that DNSBL (or have | |
| 414 nameservers that are on that list), and the host name is not on the | |
| 415 <configurable> ignore list, the mail is rejected. | |
| 416 </para> | |
| 417 <para> | |
| 418 If the content uribl DNSBL is defined, and any of those host names are | |
| 419 on that DNSBL, and the host name is not on the <configurable> | |
| 420 ignore list, the mail is rejected. | |
| 421 </para> | |
| 422 <para> | |
|
167
9b129ed78d7d
actually use spamassassin result, allow build without spam assassin, only call it if some recipient needs it.
carl
parents:
164
diff
changeset
|
423 If any non-whitelisted recipient has a filtering context with a non-zero |
|
9b129ed78d7d
actually use spamassassin result, allow build without spam assassin, only call it if some recipient needs it.
carl
parents:
164
diff
changeset
|
424 spamassassin limit, then the message is passed thru spamassassin (via |
|
9b129ed78d7d
actually use spamassassin result, allow build without spam assassin, only call it if some recipient needs it.
carl
parents:
164
diff
changeset
|
425 spamc), and the message is rejected for those recipients with spamassassin |
|
9b129ed78d7d
actually use spamassassin result, allow build without spam assassin, only call it if some recipient needs it.
carl
parents:
164
diff
changeset
|
426 limits less than the resulting spamassassin score. |
| 163 | 427 </para> |
| 428 <para> | |
| 119 | 429 We also scan for excessive bad html tags, and if a <configurable> |
| 430 limit is exceeded, the mail is rejected. | |
| 108 | 431 </para> |
| 432 </refsect1> | |
| 94 | 433 |
| 111 | 434 <refsect1 id='access.1'> |
| 108 | 435 <title>Sendmail access vs. DNSBL</title> |
| 436 <para> | |
| 437 With the standard sendmail.mc dnsbl FEATURE, the dnsbl checks may be | |
| 438 suppressed by entries in the /etc/mail/access database. For example, | |
| 439 suppose you control a /18 of address space, and have allocated some /24s | |
| 440 to some clients. You have access entries like | |
| 111 | 441 <literallayout class="monospaced"><![CDATA[ |
| 442 192.168.4 OK | |
| 443 192.168.17 OK]]></literallayout> | |
| 108 | 444 </para> |
| 445 <para> | |
| 446 to allow those clients to smarthost thru your mail server. Now if one | |
| 447 of those clients happens get infected with a virus that turns a machine | |
| 448 into an open proxy, and their 192.168.4.45 lands on the SBL-XBL, you | |
| 449 will still wind up allowing that infected machine to smarthost thru your | |
| 450 mail servers. | |
| 451 </para> | |
| 452 <para> | |
| 453 With this DNSBL milter, the sendmail access database cannot override the | |
| 454 dnsbl checks, so that machine won't be able to send mail to or thru your | |
| 455 smarthost mail server (unless the virus/proxy can use smtp-auth). | |
| 456 </para> | |
| 457 <para> | |
| 458 Using the standard sendmail features, you would add access entries to | |
| 459 allow hosts on your local network to relay thru your mail server. Those | |
| 460 OK entries in the sendmail access database will override all the dnsbl | |
| 461 checks. With this DNSBL milter, you will need to have the local users | |
| 462 authenticate with smtp-auth to get the same effect. You might find | |
| 463 <ulink | |
| 464 url="http://www.ists.dartmouth.edu/classroom/sendmail-ssl-how-to.php"> | |
| 465 these directions</ulink> helpful for setting up smtp-auth if you are on | |
| 466 RH Linux. | |
| 467 </para> | |
| 468 </refsect1> | |
| 94 | 469 |
| 111 | 470 <refsect1 id='performance.1'> |
| 108 | 471 <title>Performance Issues</title> |
| 472 <para> | |
| 473 Consider a high volume high performance machine running sendmail. Each | |
| 474 sendmail process can do its own dns resolution. Typically, such dns | |
| 475 resolver libraries are not thread safe, and so must be protected by some | |
| 476 sort of mutex in a threaded environment. When we add a milter to | |
| 477 sendmail, we now have a collection of sendmail processes, and a | |
| 478 collection of milter threads. | |
| 479 </para> | |
| 480 <para> | |
| 481 We will be doing a lot of dns lookups per mail message, and at least | |
| 482 some of those will take many tens of seconds. If all this dns work is | |
| 483 serialized inside the milter, we have an upper limit of about 25K mail | |
| 484 messages per day. That is clearly not sufficient for many sites. | |
| 485 </para> | |
| 486 <para> | |
| 487 Since we want to do parallel dns resolution across those milter threads, | |
| 488 we add another collection of dns resolver processes. Each sendmail | |
| 489 process is talking to a milter thread over a socket, and each milter | |
| 490 thread is talking to a dns resolver process over another socket. | |
| 491 </para> | |
| 492 <para> | |
| 493 Suppose we are processing 20 messages per second, and each message | |
| 494 requires 20 seconds of dns work. Then we will have 400 sendmail | |
| 495 processes, 400 milter threads, and 400 dns resolver processes. Of | |
| 496 course that steady state is very unlikely to happen. | |
| 497 </para> | |
| 498 </refsect1> | |
| 94 | 499 |
| 500 | |
| 111 | 501 <refsect1 id='rejected.1'> |
| 108 | 502 <title>Rejected Ideas</title> |
| 503 <para> | |
| 504 The following ideas have been considered and rejected. | |
| 505 </para> | |
| 506 <para> | |
| 111 | 507 Add max_recipients setting to the context configuration. Recipients in |
| 508 excess of that limit will be rejected, and all the non-whitelisted | |
| 509 recipients will be removed. Current spammers *very* rarely send more | |
| 510 than ten recipients in a single smtp transaction, so this won't stop any | |
| 108 | 511 significant amount of spam. |
| 512 </para> | |
| 513 <para> | |
| 514 Add poison addresses to the configuration. If any recipient is | |
| 515 poison, all recipients are rejected even if they would be whitelisted, | |
| 516 and the data is rejected if sent. I have a collection of spam trap | |
| 517 addresses that would be suitable for such use. Based on my log files, | |
| 518 any mail to those spam trap addresses is rejected based on either dnsbl | |
| 519 lookups or the DCC. So this won't result in blocking any additional | |
| 520 spam. | |
| 521 </para> | |
| 522 <para> | |
| 523 Add an option to only allow one recipient if the return path is | |
| 524 empty. Based on my log files, there is no mail that violates this | |
| 525 check. | |
| 526 </para> | |
| 527 <para> | |
| 528 Reject the mail if the envelope from domain name contains any MX | |
| 529 records pointing to 127.0.0.0/8. I don't see any significant amount of | |
| 530 spam sent with such domain names. | |
| 531 </para> | |
| 532 </refsect1> | |
| 94 | 533 |
| 108 | 534 <refsect1 id='todo.1'> |
| 535 <title>TODO</title> | |
| 536 <para> | |
| 537 The following ideas are under consideration. | |
| 538 </para> | |
| 539 <para> | |
| 115 | 540 Look for href="hostname/path" strings that are missing the required |
| 541 http:// protocol header. Such references are still clickable in common | |
| 542 mail software. | |
| 543 </para> | |
| 108 | 544 </refsect1> |
| 94 | 545 |
| 111 | 546 <refsect1 id='copyright.1'> |
| 108 | 547 <title>Copyright</title> |
| 548 <para> | |
| 163 | 549 Copyright (C) 2007 by 510 Software Group <carl@five-ten-sg.com> |
| 108 | 550 </para> |
| 551 <para> | |
| 552 This program is free software; you can redistribute it and/or modify it | |
| 553 under the terms of the GNU General Public License as published by the | |
| 160 | 554 Free Software Foundation; either version 3, or (at your option) any |
| 108 | 555 later version. |
| 556 </para> | |
| 557 <para> | |
| 558 You should have received a copy of the GNU General Public License along | |
| 559 with this program; see the file COPYING. If not, please write to the | |
| 560 Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. | |
| 561 </para> | |
| 562 </refsect1> | |
| 94 | 563 |
| 111 | 564 <refsect1 id='version.1'> |
|
201
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
565 <title>Version</title> |
| 108 | 566 <para> |
|
201
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
567 @VERSION@ |
| 108 | 568 </para> |
| 569 </refsect1> | |
| 570 </refentry> | |
| 571 | |
| 572 | |
| 573 <refentry id="@PACKAGE@.conf.5"> | |
| 574 <refentryinfo> | |
|
201
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
575 <date>2008-02-02</date> |
| 108 | 576 </refentryinfo> |
| 94 | 577 |
| 108 | 578 <refmeta> |
| 579 <refentrytitle>@PACKAGE@.conf</refentrytitle> | |
| 580 <manvolnum>5</manvolnum> | |
| 581 <refmiscinfo>@PACKAGE@ @VERSION@</refmiscinfo> | |
| 582 </refmeta> | |
| 94 | 583 |
| 108 | 584 <refnamediv id='name.5'> |
| 585 <refname>@PACKAGE@.conf</refname> | |
| 111 | 586 <refpurpose>configuration file for @PACKAGE@ sendmail milter</refpurpose> |
| 108 | 587 </refnamediv> |
| 588 | |
| 589 <refsynopsisdiv id='synopsis.5'> | |
| 590 <title>Synopsis</title> | |
| 591 <cmdsynopsis> | |
| 592 <command>@PACKAGE@.conf</command> | |
| 593 </cmdsynopsis> | |
| 594 </refsynopsisdiv> | |
| 94 | 595 |
| 108 | 596 <refsect1 id='description.5'> |
| 597 <title>Description</title> | |
| 598 <para>The <command>@PACKAGE@.conf</command> configuration file is | |
|
148
9330b8d6a56b
add documentation fixes, allow env_from target of inherit
carl
parents:
144
diff
changeset
|
599 specified by this partial bnf description. Comments start with // |
|
9330b8d6a56b
add documentation fixes, allow env_from target of inherit
carl
parents:
144
diff
changeset
|
600 or # and extend to the end of the line. To include the contents |
|
9330b8d6a56b
add documentation fixes, allow env_from target of inherit
carl
parents:
144
diff
changeset
|
601 of some file verbatim in the dnsbl.conf file, use |
|
9330b8d6a56b
add documentation fixes, allow env_from target of inherit
carl
parents:
144
diff
changeset
|
602 <literallayout class="monospaced"><![CDATA[include "<file>";]]></literallayout> |
|
9330b8d6a56b
add documentation fixes, allow env_from target of inherit
carl
parents:
144
diff
changeset
|
603 </para> |
| 108 | 604 |
| 605 <literallayout class="monospaced"><![CDATA[ | |
| 606 CONFIG = {CONTEXT ";"}+ | |
| 607 CONTEXT = "context" NAME "{" {STATEMENT}+ "}" | |
| 168 | 608 STATEMENT = (DNSBL | DNSBLLIST | CONTENT | ENV-TO | VERIFY | GENERIC |
| 609 | AUTOWHITE | CONTEXT | ENV-FROM | RATE-LIMIT) ";" | |
| 108 | 610 |
| 124 | 611 DNSBL = "dnsbl" NAME DNSPREFIX ERROR-MSG1 |
| 108 | 612 |
| 613 DNSBLLIST = "dnsbl_list" {NAME}+ | |
| 94 | 614 |
| 108 | 615 CONTENT = "content" ("on" | "off") "{" {CONTENT-ST}+ "}" |
| 178 | 616 CONTENT-ST = (FILTER | URIBL | IGNORE | TLD | CCTLD | HTML-TAGS | |
| 617 HTML-LIMIT | HOST-LIMIT | SPAMASS | REQUIRE | DCCGREY | | |
| 618 DCCBULK) ";" | |
| 124 | 619 FILTER = "filter" DNSPREFIX ERROR-MSG2 |
| 620 URIBL = "uribl" DNSPREFIX ERROR-MSG3 | |
| 108 | 621 IGNORE = "ignore" "{" {HOSTNAME [";"]}+ "}" |
| 622 TLD = "tld" "{" {TLD [";"]}+ "}" | |
| 119 | 623 CCTLD = "cctld" "{" {TLD [";"]}+ "}" |
| 108 | 624 HTML-TAGS = "html_tags" "{" {HTMLTAG [";"]}+ "}" |
| 124 | 625 ERROR-MSG1 = string containing exactly two %s replacement tokens |
| 626 both are replaced with the client ip address | |
| 627 ERROR-MSG2 = string containing exactly two %s replacement tokens | |
| 628 the first is replaced with the hostname, and the second | |
| 629 is replaced with the ip address | |
| 630 ERROR-MSG3 = string containing exactly two %s replacement tokens | |
| 631 both are replaced with the hostname | |
| 108 | 632 |
| 633 HTML-LIMIT = "html_limit" ("on" INTEGER ERROR-MSG | "off") | |
| 634 | |
| 111 | 635 HOST-LIMIT = "host_limit" ("on" INTEGER ERROR-MSG | "off" | |
| 636 "soft" INTEGER) | |
| 178 | 637 SPAMASS = "spamassassin" INTEGER |
| 638 REQUIRE = "require_match" ("yes" | "no") | |
| 639 DCCGREY = "dcc_greylist" ("yes" | "no") | |
| 640 DCCBULK = "dcc_bulk_threshold" (INTEGER | "many" | "off") | |
| 94 | 641 |
| 108 | 642 ENV-TO = "env_to" "{" {(TO-ADDR | DCC-TO)}+ "}" |
| 643 TO-ADDR = ADDRESS [";"] | |
| 644 DCC-TO = "dcc_to" ("ok" | "many") "{" DCCINCLUDEFILE "}" ";" | |
| 645 | |
| 646 VERIFY = "verify" HOSTNAME ";" | |
| 168 | 647 GENERIC = "generic" REGULAREXPRESSION ERROR-MSG4 ";" |
| 648 ERROR-MSG4 = string containing exactly one %s replacement token | |
| 649 which is replaced with the client name | |
| 153 | 650 AUTOWHITE = "autowhite" DAYS FILENAME ";" |
| 108 | 651 |
| 652 ENV_FROM = "env_from" [DEFAULT] "{" {(FROM-ADDR | DCC-FROM)}+ "}" | |
| 653 FROM-ADDR = ADDRESS VALUE [";"] | |
| 654 DCC-FROM = "dcc_from" "{" DCCINCLUDEFILE "}" ";" | |
| 136 | 655 |
| 140 | 656 RATE-LIMIT = "rate_limit" [DEFAULTLIMIT] "{" (RATE)+ "}" |
| 136 | 657 RATE = USER LIMIT [";"] |
| 658 | |
| 108 | 659 DEFAULT = ("white" | "black" | "unknown" | "inherit" | "") |
| 660 ADDRESS = (USER@ | DOMAIN | USER@DOMAIN) | |
|
148
9330b8d6a56b
add documentation fixes, allow env_from target of inherit
carl
parents:
144
diff
changeset
|
661 VALUE = ("white" | "black" | "unknown" | "inherit" | CHILD-CONTEXT-NAME)]]></literallayout> |
| 108 | 662 </refsect1> |
| 94 | 663 |
| 108 | 664 <refsect1 id='sample.5'> |
| 665 <title>Sample</title> | |
| 666 <literallayout class="monospaced"><![CDATA[ | |
| 127 | 667 context main-default { |
| 668 // outbound dnsbl filtering to catch our own customers that end up on the sbl | |
| 669 dnsbl sbl sbl-xbl.spamhaus.org "Mail from %s rejected - sbl; see http://www.spamhaus.org/query/bl?ip=%s"; | |
| 174 | 670 dnsbl_list sbl; |
| 127 | 671 |
| 672 // outbound content filtering to prevent our own customers from sending spam | |
| 673 content on { | |
| 674 filter sbl-xbl.spamhaus.org "Mail containing %s rejected - sbl; see http://www.spamhaus.org/query/bl?ip=%s"; | |
| 675 uribl multi.surbl.org "Mail containing %s rejected - surbl; see http://www.rulesemporium.com/cgi-bin/uribl.cgi?bl0=1&domain0=%s"; | |
| 676 #uribl black.uribl.com "Mail containing %s rejected - uribl; see http://l.uribl.com/?d=%s"; | |
| 677 ignore { include "hosts-ignore.conf"; }; | |
| 678 tld { include "tld.conf"; }; | |
| 679 cctld { include "cctld.conf"; }; | |
| 680 html_tags { include "html-tags.conf"; }; | |
| 681 html_limit on 20 "Mail containing excessive bad html tags rejected"; | |
| 682 html_limit off; | |
| 683 host_limit on 20 "Mail containing excessive host names rejected"; | |
| 684 host_limit soft 20; | |
| 178 | 685 spamassassin 4; |
| 686 require_match yes; | |
| 687 dcc_greylist yes; | |
| 688 dcc_bulk_threshold 50; | |
| 127 | 689 }; |
| 690 | |
| 691 // backscatter prevention - don't send bounces for mail that we accepted but could not forward | |
| 692 // we only send bounces to our own customers | |
| 693 env_from unknown { | |
| 694 "<>" black; | |
| 695 }; | |
| 136 | 696 |
|
144
31ff00ea6bfb
allow parent/child to share a fully qualified env_to address
carl
parents:
140
diff
changeset
|
697 // hourly recipient rate limit by smtp auth client id |
| 140 | 698 rate_limit 30 { // default |
| 171 | 699 #fred 100; // override default limits |
| 700 #joe 10; // "" | |
| 136 | 701 }; |
| 127 | 702 }; |
| 703 | |
| 171 | 704 context main { |
| 705 dnsbl localp partial.blackholes.five-ten-sg.com "Mail from %s rejected - local; see http://www.five-ten-sg.com/blackhole.php?%s"; | |
| 108 | 706 dnsbl local blackholes.five-ten-sg.com "Mail from %s rejected - local; see http://www.five-ten-sg.com/blackhole.php?%s"; |
| 174 | 707 dnsbl sbl zen.spamhaus.org "Mail from %s rejected - sbl; see http://www.spamhaus.org/query/bl?ip=%s"; |
| 108 | 708 dnsbl xbl xbl.spamhaus.org "Mail from %s rejected - xbl; see http://www.spamhaus.org/query/bl?ip=%s"; |
| 171 | 709 dnsbl_list local sbl; |
| 94 | 710 |
| 108 | 711 content on { |
| 712 filter sbl-xbl.spamhaus.org "Mail containing %s rejected - sbl; see http://www.spamhaus.org/query/bl?ip=%s"; | |
| 122 | 713 uribl multi.surbl.org "Mail containing %s rejected - surbl; see http://www.rulesemporium.com/cgi-bin/uribl.cgi?bl0=1&domain0=%s"; |
| 119 | 714 #uribl black.uribl.com "Mail containing %s rejected - uribl; see http://l.uribl.com/?d=%s"; |
| 108 | 715 ignore { include "hosts-ignore.conf"; }; |
| 716 tld { include "tld.conf"; }; | |
| 119 | 717 cctld { include "cctld.conf"; }; |
| 108 | 718 html_tags { include "html-tags.conf"; }; |
| 719 html_limit off; | |
| 720 host_limit soft 20; | |
| 178 | 721 spamassassin 5; |
| 722 require_match yes; | |
| 723 dcc_greylist yes; | |
| 724 dcc_bulk_threshold 20; | |
| 108 | 725 }; |
| 94 | 726 |
|
175
e726e1a61ef9
allow manual whitelisting with stamp 1 to remove a whitelist entry
carl
parents:
174
diff
changeset
|
727 generic "(^|[.-])(ppp|h|host)?([0-9]{1,3}[.-](Red-|dynamic[.-])?){4}" |
| 171 | 728 "your mail server %s seems to have a generic name"; |
| 729 | |
| 108 | 730 env_to { |
| 171 | 731 # !! replace this with your domain names |
| 108 | 732 # child contexts are not allowed to specify recipient addresses outside these domains |
| 179 | 733 # if this is a backup-mx, you need to include here domains for which you relay to the primary mx |
| 174 | 734 include "/etc/mail/local-host-names"; |
| 108 | 735 }; |
| 94 | 736 |
| 108 | 737 context whitelist { |
| 738 content off {}; | |
| 739 env_to { | |
| 171 | 740 # dcc_to ok { include "/var/dcc/whitecommon"; }; |
| 108 | 741 }; |
| 742 env_from white {}; # white forces all unmatched from addresses (everyone in this case) to be whitelisted | |
| 743 # so all mail TO these env_to addresses is accepted | |
| 744 }; | |
| 94 | 745 |
| 171 | 746 context abuse { |
| 747 dnsbl_list xbl; | |
| 748 content off {}; | |
| 174 | 749 generic "^$ " " "; # regex cannot match, to disable generic rdns rejects |
| 171 | 750 env_to { |
| 751 abuse@ # no content filtering on abuse reports | |
| 752 postmaster@ # "" | |
| 753 }; | |
| 754 env_from unknown {}; # ignore all parent white/black listing | |
| 755 }; | |
| 756 | |
| 108 | 757 context minimal { |
| 171 | 758 dnsbl_list sbl; |
| 178 | 759 content on { |
| 760 spamassassin 10; | |
| 761 dcc_bulk_threshold many; | |
| 762 }; | |
| 171 | 763 generic "^$ " " "; # regex cannot match, to disable generic rdns rejects |
| 108 | 764 env_to { |
| 765 }; | |
| 766 }; | |
| 94 | 767 |
| 108 | 768 context blacklist { |
| 769 env_to { | |
| 171 | 770 # dcc_to many { include "/var/dcc/whitecommon"; }; |
| 108 | 771 }; |
| 772 env_from black {}; # black forces all unmatched from addresses (everyone in this case) to be blacklisted | |
| 773 # so all mail TO these env_to addresses is rejected | |
| 774 }; | |
| 94 | 775 |
| 171 | 776 env_from unknown { |
| 777 abuse@ abuse; # replies to abuse reports use the abuse context | |
| 778 # dcc_from { include "/var/dcc/whitecommon"; }; | |
| 108 | 779 }; |
| 780 | |
| 171 | 781 autowhite 90 "autowhite/my-auto-whitelist"; |
| 782 # install should create /etc/dnsbl/autowhite writable by userid dnsbl | |
| 108 | 783 };]]></literallayout> |
| 784 </refsect1> | |
| 94 | 785 |
| 111 | 786 <refsect1 id='version.5'> |
|
201
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
787 <title>Version</title> |
| 108 | 788 <para> |
|
201
752d4315675c
add reference to mercurial repository in the documentation
Carl Byington <carl@five-ten-sg.com>
parents:
187
diff
changeset
|
789 @VERSION@ |
| 108 | 790 </para> |
| 791 </refsect1> | |
| 792 | |
| 793 </refentry> | |
| 794 </reference> |