-------------------------------------------------------------- The exiscan-acl patch for exim4 - Documentation -------------------------------------------------------------- (c) Tom Kistner 2003-???? License: GPL The exiscan-acl patch adds content scanning to the exim4 ACL system. It supports the following scanning features: - MIME ACL that is called for all MIME parts in incoming MIME messages. - Antivirus using 3rd party scanners. - Anti-spam using SpamAssassin. - Anti-spam using Brightmail Antispam. - SPF support via the libspf2 library - SRS support via the libsrs_alt library - Regular expression match against headers, bodies, raw MIME parts and decoded MIME parts. These features are hooked into exim by extending exim's ACL system. The patch adds expansion variables and ACL conditions. These conditions are designed to be used in the acl_smtp_data ACL. It is run when the sending host has completed the DATA phase and is waiting for our final response to his end-of-data marker. This allows us to reject messages containing unwanted content at that stage. Support for Brightmail AntiSpam requires special compile-time flags. Please refer to chapter 7 for details. The default exim configure file contains commented configuration examples for some features of exiscan-acl. 0. Overall concept / Overview -------------------------------------------------------------- The exiscan-acl patch extends Exims with mechanisms to deal with the message body content. Most of these additions affect the ACL system. The exiscan patch adds - A new ACL, called 'acl_smtp_mime' (Please see detailed chapter on this one below). - ACL conditions and modifiers o malware (attach 3rd party virus/malware scanner) o spam (attach SpamAssassin) o regex (match regex against message, linewise) o decode (decode MIME part to disk) o mime_regex (match regex against decoded MIME part) o control = fakereject (reject but really accept a message) o spf (do SPF checks) - expansion variables (see chapters below for names and explanations) - configuration options in section 1 of Exim's configure file. o av_scanner (type and options of the AV scanner) o spamd_address (network address / socket of spamd daemon). All facilites work on a MBOX copy of the message that is temporarily spooled up in a file called: /scan//.eml The .eml extension is a friendly hint to virus scanners that they can expect an MBOX-like structure inside that file. The file is only spooled up once, when the first exiscan facility is called. Subsequent calls to exiscan conditions will just open the file again. The directory is recursively removed when the acl_smtp_data has finished running. When the MIME ACL decodes files, they will be put into that same folder by default. 1. The acl_smtp_mime MIME ACL -------------------------------------------------------------- Note: if you are not familiar with exims ACL system, please go read the documentation on it, otherwise this chapter will not make much sense to you. Here are the facts on acl_smtp_mime: - It is called once for each MIME part of a message, including multipart types, in the sequence of their position in the message. - It is called just before the acl_smtp_data ACL. They share a result code (the one assed to the remote system after DATA). When a call to acl_smtp_mime does not yield "accept", ACL processing is aborted and the respective result code is sent to the remote mailer. This means that the acl_smtp_data is NOT called any more. - It is ONLY called if the message has a MIME-Version header. - MIME parts will NOT be dumped to disk by default, you have to call the "decode" condition to do that (see further below). - For RFC822 attachments (these are messages attached to messages, with a content-type of 'message/rfc822'), the ACL is called again in the same manner as for the "primary" message, only that the $mime_is_rfc822 expansion variable is set (see below). These messages are always decoded to disk before being checked, but the files are unlinked once the check is done. To activate acl_smtp_mime, you need to add assign it the name of an ACL entry in section 1 of the config file, and then write that ACL in the ACL section, like: /* --------------- # -- section 1 ---- [ ... ] acl_smtp_mime = my_mime_acl [ ... ] # -- acl section ---- begin acl [ ... ] my_mime_acl: < ACL logic > [ ... ] ---------------- */ The following list describes all expansion variables that are available in the MIME ACL: $mime_content_type ------------------ A very important variable. If the MIME part has a "Content -Type:" header, this variable will contain its value, lowercased, and WITHOUT any options (like "name" or "charset", see below for these). Here are some examples of popular MIME types, as they may appear in this variable: text/plain text/html application/octet-stream image/jpeg audio/midi If the MIME part has no "Content-Type:" header, this variable is the empty string. $mime_filename -------------- Another important variable, possibly the most important one. It contains a proposed filename for an attachment, if one was found in either the "Content-Type:" or "Content -Disposition" headers. The filename will be RFC2047 decoded, however NO additional sanity checks are done. See instructions on "decode" further below. If no filename was found, this variable is the empty string. $mime_charset ------------- Contains the charset identifier, if one was found in the "Content-Type:" header. Examples for charset identifiers are us-ascii gb2312 (Chinese) iso-8859-1 Please note that this value will NOT be normalized, so you should do matches case-insensitively. $mime_boundary -------------- If the current part is a multipart (see $mime_is_multipart) below, it SHOULD have a boundary string. It is stored in this variable. If the current part has no boundary parameter in the "Content-Type:" header, this variable contains the empty string. $mime_content_disposition ------------------------- Contains the normalized content of the "Content -Disposition:" header. You can expect strings like "attachment" or "inline" here. $mime_content_transfer_encoding ------------------------------- Contains the normalized content of the "Content -transfer-encoding:" header. This is a symbolic name for an encoding type. Typical values are "base64" and "quoted -printable". $mime_content_id ---------------- Contains the normalized content of the "Content -ID:" header. This is a unique ID that can be used to reference a part from another part. $mime_content_description ------------------------- Contains the normalized content of the "Content -Description:" header. It can contain a human-readable description of the parts content. Some implementations will repeat the filename for attachments here, but they are usually only used for display purposes. $mime_part_count ---------------- This is a counter that is raised for each processed MIME part. It starts at zero for the very first part (which is usually a multipart). The counter is per-message, so it is reset when processing RFC822 attachments (see $mime_is_rfc822). The counter stays set after acl_smtp_mime is complete, so you can use it in the DATA ACL to determine the number of MIME parts of a message. For non-MIME messages, this variable will contain the value -1. $mime_is_multipart ------------------ A "helper" flag that is true (1) when the current part has the main type "multipart", for example "multipart/alternative" or "multipart/mixed". Since multipart entities only serve as containers for other parts, you may not want to carry out specific actions on them. $mime_is_coverletter -------------------- This flag attempts to differentiate the "cover letter" of an e-mail from attached data. It can be used to clamp down on "flashy" or unneccessarily encoded content in the coverletter, while not restricting attachments at all. It returns 1 (true) for a MIME part believed to be part of the coverletter, 0 (false) for an attachment. At present, the algorithm is as follows: 1. The outermost MIME part of a message always coverletter. 2. If a multipart/alternative or multipart/related MIME part is coverletter, so are all MIME subparts within that multipart. 3. If any other multipart is coverletter, the first subpart is coverletter, the rest are attachments. 4. All parts contained within an attachment multipart are attachments. As an example, the following will ban "HTML mail" (including that sent with alternative plain text), while allowing HTML files to be attached: /* ---------------------- deny message = HTML mail is not accepted here condition = $mime_is_coverletter condition = ${if eq{$mime_content_type}{text/html}{1}{0}} ----------------------- */ $mime_is_rfc822 --------------- This flag is true (1) if the current part is NOT a part of the checked message itself, but part of an attached message. Attached message decoding is fully recursive. $mime_decoded_filename ---------------------- This variable is only set after the "decode" condition (see below) has been successfully run. It contains the full path and file name of the file containing the decoded data. $mime_content_size ------------------ This variable is only set after the "decode" condition (see below) has been successfully run. It contains the size of the decoded part in kilobytes(!). The size is always rounded up to full kilobytes, so only a completely empty part will have a mime_content_size of 0. The expansion variables only reflect the content of the MIME headers for each part. To actually decode the part to disk, you can use the "decode" condition. The general syntax is decode = [//] The right hand side is expanded before use. After expansion, the value can - be '0' or 'false', in which case no decoding is done. - be the string 'default'. In that case, the file will be put in the temporary "default" directory /scan// with a sequential file name, consisting of the message id and a sequence number. The full path and name is available in $mime_decoded_filename after decoding. - start with a slash. If the full name is an existing directory, it will be used as a replacement for the "default" directory. The filename will then also be sequentially assigned. If the name does not exist, it will be used as the full path and file name. - not start with a slash. It will then be used as the filename, and the default path will be used. You can easily decode a file with its original, proposed filename using "decode = $mime_filename". However, you should keep in mind that $mime_filename might contain anything. If you place files outside of the default path, they will not be automatically unlinked. The MIME ACL also supports the regex= and mime_regex= conditions. You can use those to match regular expressions against raw and decoded MIME parts, respectively. Read the next section for more information on these conditions. 2. Match message or MIME parts against regular expressions -------------------------------------------------------------- The "regex" condition takes one or more regular expressions as arguments and matches them against the full message (when called in the DATA ACL) or a raw MIME part (when called in the MIME ACL). The "regex" condition matches linewise, with a maximum line length of 32k characters. That means you can't have multiline matches with the "regex" condition. The "mime_regex" can only be called in the MIME ACL. It matches up to 32k of decoded content (the whole content at once, not linewise). If the part has not been decoded with the "decode" condition earlier in the ACL, it is decoded automatically when "mime_regex" is executed (using default path and filename values). If the decoded data is larger than 32k, only the first 32k characters will be matched. The regular expressions are passed as a colon-separated list. To include a literal colon, you must double it. Since the whole right-hand side string is expanded before being used, you must also escape dollar ($) signs with backslashes. Here is a simple example: /* ---------------------- deny message = contains blacklisted regex ($regex_match_string) regex = [Mm]ortgage : URGENT BUSINESS PROPOSAL ----------------------- */ The conditions returns true if one of the regular expressions has matched. The $regex_match_string expansion variable is then set up and contains the matching regular expression. Warning: With large messages, these conditions can be fairly CPU-intensive. 3. Antispam measures with SpamAssassin -------------------------------------------------------------- The "spam" ACL condition calls SpamAssassin's "spamd" daemon to get a spam-score and a report for the message. You must first install SpamAssassin. You can get it at http://www.spamassassin.org, or, if you have a working Perl installation, you can use CPAN by calling perl -MCPAN -e 'install Mail::SpamAssassin' SpamAssassin has its own set of configuration files. Please review its documentation to see how you can tweak it. The default installation should work nicely, however. After having installed and configured SpamAssassin, start the "spamd" daemon. By default, it listens on 127.0.0.1, TCP port 783. If you use another host or port for spamd, you must set the spamd_address option in Section 1 of the exim configuration as follows (example): spamd_address = 127.0.0.1 783 As of version 2.60, spamd also supports communication over UNIX sockets. If you want to use these, supply spamd_address with an absolute file name instead of a address/port pair, like: spamd_address = /var/run/spamd_socket If you use the above mentioned default, you do NOT need to set this option. You can also have multiple spamd servers to improve scalability. These can also reside on other hardware reachable over the network. To specify multiple spamd servers, just put multiple address/port pairs in the spamd_address option, separated with colons: spamd_address = 192.168.2.10 783 : 192.168.2.11 783 : 192.168.2.12 783 Up to 32 spamd servers are supported. The servers will be queried in a random fashion. When a server fails to respond to the connection attempt, all other servers are tried until one succeeds. If no server responds, the "spam" condition will return DEFER. IMPORTANT: It is not possible to use the UNIX socket connection method with multiple spamd servers. To use the antispam facility, put the "spam" condition in a DATA ACL block. Here is a very simple example: /* --------------- deny message = This message was classified as SPAM spam = joe ---------------- */ On the right-hand side of the spam condition, you can put the username that SpamAssassin should scan for. That allows you to use per-domain or per-user antispam profiles. The right-hand side is expanded before being used, so you can put lookups or conditions there. When the right-hand side evaluates to "0" or "false", no scanning will be done and the condition will fail immediately. If you do not want to scan for a particular user, but rather use the SpamAssassin system-wide default profile, you can scan for an unknown user, or simply use "nobody". The "spam" condition will return true if the threshold specified in the user's SpamAssassin profile has been matched or exceeded. If you want to use the spam condition for its side effects (see the variables below), you can make it always return "true" by appending ":true" to the username. When the condition is run, it sets up the following expansion variables: $spam_score The spam score of the message, for example "3.4" or "30.5". This is useful for inclusion in log or reject messages. $spam_score_int The spam score of the message, multiplied by ten, as an integer value. For example "34" or "305". This is useful for numeric comparisons in conditions. See further below for a more complicated example. This variable is special, since it is written to the spool file, so it can be used during the whole life of the message on your exim system, even in routers or transports. $spam_bar A string consisting of a number of '+' or '-' characters, representing the spam_score value. A spam score of "4.4" would have a spam_bar of '++++'. This is useful for inclusion in warning headers, since MUAs can match on such strings. $spam_report A multiline text table, containing the full SpamAssassin report for the message. Useful for inclusion in headers or reject messages. The spam condition caches its results. If you call it again with the same user name, it will not really scan again, but rather return the same values as before. The spam condition will return DEFER if there is any error while running the message through SpamAssassin. If you want to treat DEFER as FAIL (to pass on to the next ACL statement block), append '/defer_ok' to the right hand side of the spam condition, like this: /* --------------- deny message = This message was classified as SPAM spam = joe/defer_ok ---------------- */ This will cause messages to be accepted even if there is a problem with spamd. Finally, here is a commented example on how to use the spam condition: /* ---------------- # put headers in all messages (no matter if spam or not) warn message = X-Spam-Score: $spam_score ($spam_bar) spam = nobody:true warn message = X-Spam-Report: $spam_report spam = nobody:true # add second subject line with *SPAM* marker when message # is over threshold warn message = Subject: *SPAM* $h_Subject spam = nobody # reject spam at high scores (> 12) deny message = This message scored $spam_score spam points. spam = nobody:true condition = ${if >{$spam_score_int}{120}{1}{0}} ----------------- */ 4. The "malware" facility Scan messages for viruses using an external virus scanner -------------------------------------------------------------- This facility lets you connect virus scanner software to exim. It supports a "generic" interface to scanners called via the shell, and specialized interfaces for "daemon" type virus scanners, who are resident in memory and thus are much faster. To use this facility, you MUST set the "av_scanner" option in section 1 of the exim config file. It specifies the scanner type to use, and any additional options it needs to run. The basic syntax is as follows: av_scanner = :::[...] The following scanner-types are supported in this release: sophie Sophie is a daemon that uses Sophos' libsavi library to scan for viruses. You can get Sophie at http://www.vanja.com/tools/sophie/. The only option for this scanner type is the path to the UNIX socket that Sophie uses for client communication. The default path is /var/run/sophie, so if you are using this, you can omit the option. Example: av_scanner = sophie:/tmp/sophie aveserver This is the scanner daemon of Kaspersky Version 5. You can get a trial version at http://www.kaspersky.com. This scanner type takes one option, which is the path to the daemon's UNIX socket. The default is "/var/run/aveserver". Example: av_scanner = aveserver:/var/run/aveserver kavdaemon This is the scanner daemon of Kaspersky Version 4. This version of the Kaspersky scanner is outdated. Please upgrade (see "aveserver" above). This scanner type takes one option, which is the path to the daemon's UNIX socket. The default is "/var/run/AvpCtl". Example: av_scanner = kavdaemon:/opt/AVP/AvpCtl clamd Another daemon type scanner, this one is GPL and free. Get it at http://clamav.elektrapro.com/. Clamd does not seem to unpack MIME containers, so it is recommended to use the demime facility with it. It takes one option: either the path and name of a UNIX socket file, or a hostname/port pair, separated by space. If unset, the default is "/tmp/clamd". Example: av_scanner = clamd:192.168.2.100 1234 or av_scanner = clamd:/opt/clamd/socket fsecure This is for the F-Secure (www.f-secure.com) scanner. It takes one argument which is the path to a UNIX socket. If omitted, it defaults to /var/run/.fsav Example: av_scanner = fsecure:/path/to/.fsav Thanks to Johan Thelmen for contributing the code for this scanner. drweb This one is for the DrWeb (http://www.sald.com/) daemon. It takes one argument, either a full path to a UNIX socket, or an IP address and port separated by whitespace. If you omit the argument, the default /usr/local/drweb/run/drwebd.sock is used. Example: av_scanner = drweb:192.168.2.20 31337 or av_scanner = drweb:/var/run/drwebd.sock Thanks to Alex Miller for contributing the code for this scanner. mksd Yet another daemon type scanner, aimed mainly at Polish users, though some parts of documentation are now avaliable in English. You can get it at http://linux.mks.com.pl/. The only option for this scanner type is the maximum number of processes used simultaneously to scan the attachments, provided that the demime facility is employed and also mksd has been run with at least the same number of child processes. You can safely omit this option, the default value is 1. Example: av_scanner = mksd:2 cmdline This is the keyword for the generic command line scanner interface. It can be used to attach virus scanners that are invoked on the shell. This scanner type takes 3 mantadory options: - full path and name of the scanner binary, with all command line options and a placeholder (%s) for the directory to scan. - A regular expression to match against the STDOUT and STDERR output of the virus scanner. If the expression matches, a virus was found. You must make absolutely sure that this expression only matches on "virus found". This is called the "trigger" expression. - Another regular expression, containing exactly ONE pair of braces, to match the name of the virus found in the scanners output. This is called the "name" expression. Example: Sophos Sweep reports a virus on a line like this: Virus 'W32/Magistr-B' found in file ./those.bat For the "trigger" expression, we just use the "found" word. For the "name" expression, we want to get the W32/Magistr-B string, so we can match for the single quotes left and right of it, resulting in the regex '(.*)' (WITH the quotes!) Altogether, this makes the configuration setting: av_scanner = cmdline:\ /path/to/sweep -all -rec -archive %s:\ found:'(.+)' When av_scanner is correcly set, you can use the "malware" condition in the DATA ACL. The condition takes a right-hand argument that is expanded before use. It can then be one of - "true", "*", or "1", in which case the message is scanned for viruses. The condition will succeed if a virus was found, or fail otherwise. This is the recommended usage. - "false" or "0", in which case no scanning is done and the condition will fail immediately. - a regular expression, in which case the message is scanned for viruses. The condition will succeed if a virus found found and its name matches the regular expression. This allows you to take special actions on certain types of viruses. When a virus was found, the condition sets up an expansion variable called $malware_name that contains the name of the virus found. You should use it in a "message" modifier that contains the error returned to the sender. The malware condition caches its results, so when you use it multiple times, the actual scanning process is only carried out once. If your virus scanner cannot unpack MIME and TNEF containers itself, you should use the demime condition prior to the malware condition. Here is a simple example: /* ---------------------- deny message = This message contains malware ($malware_name) demime = * malware = * ---------------------- */ Like with the "spam" condition, you can append '/defer_ok' to the malware condition to accept messages even if there is a problem with the virus scanner, like this: /* ---------------------- deny message = This message contains malware ($malware_name) demime = * malware = */defer_ok ---------------------- */ ADVANCED TIP: When the av_scanner option starts with a dollar ($) sign, it is expanded before being used. This is useful if you want to use multiple scanners. You can then set /* ---------------------- av_scanner = $acl_m0 ---------------------- */ and use these ACL blocks to scan with both sophie and aveserver scanners: /* ---------------------- deny message = This message contains malware ($malware_name) set acl_m0 = sophie malware = * deny message = This message contains malware ($malware_name) set acl_m0 = aveserver malware = * ---------------------- */ However, when av_scanner is expanded, the result caching of the malware condition is not used, so each malware condition call results in a new scan of the message. 5. The "demime" facility MIME unpacking, sanity checking and file extension blocking -------------------------------------------------------------- * This facility provides a simpler interface to MIME decoding * than the MIME ACL functionality. It is kept in exiscan for * backward compatability. The demime facility unpacks MIME containers in the message. It detects errors in MIME containers and can match file extensions found in the message against a list. Using this facility will produce additional files in the temporary scan directory that contain the unpacked MIME parts of the message. If you do antivirus scanning, it is recommened to use the "demime" condition before the antivirus ("malware") condition. The condition name of this facility is "demime". On the right hand side, you can pass a colon-separated list of file extensions that it should match against. If one of the file extensions is found, the condition will return "OK" (or "true"), otherwise it will return FAIL (or "false"). If there was any TEMPORARY error while demimeing (mostly "disk full"), the condition will return DEFER, and the message will be temporarily rejected. The right-hand side gets "expanded" before being treated as a list, so you can have conditions and lookups there. If it expands to an empty string, "false", or zero ("0"), no demimeing is done and the conditions returns FALSE. A short example: /* ------------ deny message = Found blacklisted file attachment demime = vbs:com:bat:pif:prf:lnk --------------- */ When the condition is run, it sets up the following expansion variables: $demime_errorlevel When an error was detected in a MIME container, this variable contains the "severity" of the error, as an integer number. The higher the value, the more severe the error. If this variable is unset or zero, no error has occured. $demime_reason When $demime_errorlevel is greater than zero, this variable contains a human -readable text string describing the MIME error that occured. $found_extension When the "demime" condition returns "true", this variable contains the file extension it has found. Both $demime_errorlevel and $demime_reason are set with the first call of the "demime" condition, and are not changed on subsequent calls. If do not want to check for any file extensions, but rather use the demime facility for unpacking or error checking purposes, just pass "*" as the right-hand side value. Here is a more elaborate example on how to use this facility: /* ----------------- # Reject messages with serious MIME container errors deny message = Found MIME error ($demime_reason). demime = * condition = ${if >{$demime_errorlevel}{2}{1}{0}} # Reject known virus spreading file extensions. # Accepting these is pretty much braindead. deny message = contains $found_extension file (blacklisted). demime = com:vbs:bat:pif:scr # Freeze .exe and .doc files. Postmaster can # examine them and eventually thaw them up. deny log_message = Another $found_extension file. demime = exe:doc control = freeze --------------------- */ 6. The "fakereject" control statement Reject a message while really accepting it. -------------------------------------------------------------- When you put "control = fakereject" in an ACL statement, the following will happen: If exim would have accepted the message, it will tell the remote host that it did not, with a message of: 550-FAKE_REJECT id=xxxxxx-xxxxxx-xx 550-Your message has been rejected but is being kept for evaluation. 550 If it was a legit message, it may still be delivered to the target recipient(s). But exim will go on to treat the message as if it had accepted it. This should be used with extreme caution, please look into the examples document for possible usage. 7. Brighmail AntiSpam (BMI) suppport -------------------------------------------------------------- Brightmail AntiSpam is a commercial package. Please see http://www.brightmail.com for more information on the product. For the sake of clarity, we'll refer to it as "BMI" from now on. 0) BMI concept and implementation overview In contrast to how spam-scanning with SpamAssassin is implemented in exiscan-acl, BMI is more suited for per -recipient scanning of messages. However, each messages is scanned only once, but multiple "verdicts" for multiple recipients can be returned from the BMI server. The exiscan implementation passes the message to the BMI server just before accepting it. It then adds the retrieved verdicts to the messages header file in the spool. These verdicts can then be queried in routers, where operation is per-recipient instead of per-message. To use BMI, you need to take the following steps: 1) Compile Exim with BMI support 2) Set up main BMI options (top section of exim config file) 3) Set up ACL control statement (ACL section of the config file) 4) Set up your routers to use BMI verdicts (routers section of the config file). 5) (Optional) Set up per-recipient opt-in information. These four steps are explained in more details below. 1) Adding support for BMI at compile time To compile with BMI support, you need to link Exim against the Brighmail client SDK, consisting of a library (libbmiclient_single.so) and a header file (bmi_api.h). You'll also need to explicitly set a flag in the Makefile to include BMI support in the Exim binary. Both can be achieved with these 2 lines in Local/Makefile: CFLAGS=-DBRIGHTMAIL -I/path/to/the/dir/with/the/includefile EXTRALIBS_EXIM=-L/path/to/the/dir/with/the/library -lbmiclient_single If you use other CFLAGS or EXTRALIBS_EXIM settings then merge the content of these lines with them. Note for BMI6.x users: You'll also have to add -lxml2_single to the EXTRALIBS_EXIM line. Users of 5.5x do not need to do this. You should also include the location of libbmiclient_single.so in your dynamic linker configuration file (usually /etc/ld.so.conf) and run "ldconfig" afterwards, or else the produced Exim binary will not be able to find the library file. 2) Setting up BMI support in the exim main configuration To enable BMI support in the main exim configuration, you should set the path to the main BMI configuration file with the "bmi_config_file" option, like this: bmi_config_file = /opt/brightmail/etc/brightmail.cfg This must go into section 1 of exims configuration file (You can put it right on top). If you omit this option, it defaults to /opt/brightmail/etc/brightmail.cfg. Note for BMI6.x users: This file is in XML format in V6.xx and its name is /opt/brightmail/etc/bmiconfig.xml. So BMI 6.x users MUST set the bmi_config_file option. 3) Set up ACL control statement To optimize performance, it makes sense only to process messages coming from remote, untrusted sources with the BMI server. To set up a messages for processing by the BMI server, you MUST set the "bmi_run" control statement in any ACL for an incoming message. You will typically do this in an "accept" block in the "acl_check_rcpt" ACL. You should use the "accept" block(s) that accept messages from remote servers for your own domain(s). Here is an example that uses the "accept" blocks from exims default configuration file: accept domains = +local_domains endpass verify = recipient control = bmi_run accept domains = +relay_to_domains endpass verify = recipient control = bmi_run If bmi_run is not set in any ACL during reception of the message, it will NOT be passed to the BMI server. 4) Setting up routers to use BMI verdicts When a message has been run through the BMI server, one or more "verdicts" are present. Different recipients can have different verdicts. Each recipient is treated individually during routing, so you can query the verdicts by recipient at that stage. From Exims view, a verdict can have the following outcomes: o deliver the message normally o deliver the message to an alternate location o do not deliver the message To query the verdict for a recipient, the implementation offers the following tools: - Boolean router preconditions. These can be used in any router. For a simple implementation of BMI, these may be all that you need. The following preconditions are available: o bmi_deliver_default This precondition is TRUE if the verdict for the recipient is to deliver the message normally. If the message has not been processed by the BMI server, this variable defaults to TRUE. o bmi_deliver_alternate This precondition is TRUE if the verdict for the recipient is to deliver the message to an alternate location. You can get the location string from the $bmi_alt_location expansion variable if you need it. See further below. If the message has not been processed by the BMI server, this variable defaults to FALSE. o bmi_dont_deliver This precondition is TRUE if the verdict for the recipient is NOT to deliver the message to the recipient. You will typically use this precondition in a top-level blackhole router, like this: # don't deliver messages handled by the BMI server bmi_blackhole: driver = redirect bmi_dont_deliver data = :blackhole: This router should be on top of all others, so messages that should not be delivered do not reach other routers at all. If the message has not been processed by the BMI server, this variable defaults to FALSE. - A list router precondition to query if rules "fired" on the message for the recipient. Its name is "bmi_rule". You use it by passing it a colon-separated list of rule numbers. You can use this condition to route messages that matched specific rules. Here is an example: # special router for BMI rule #5, #8 and #11 bmi_rule_redirect: driver = redirect bmi_rule = 5:8:11 data = postmaster@mydomain.com - Expansion variables. Several expansion variables are set during routing. You can use them in custom router conditions, for example. The following variables are available: o $bmi_base64_verdict This variable will contain the BASE64 encoded verdict for the recipient being routed. You can use it to add a header to messages for tracking purposes, for example: localuser: driver = accept check_local_user headers_add = X-Brightmail-Verdict: $bmi_base64_verdict transport = local_delivery If there is no verdict available for the recipient being routed, this variable contains the empty string. o $bmi_base64_tracker_verdict This variable will contain a BASE64 encoded subset of the verdict information concerning the "rules" that fired on the message. You can add this string to a header, commonly named "X-Brightmail-Tracker". Example: localuser: driver = accept check_local_user headers_add = X-Brightmail-Tracker: $bmi_base64_tracker_verdict transport = local_delivery If there is no verdict available for the recipient being routed, this variable contains the empty string. o $bmi_alt_location If the verdict is to redirect the message to an alternate location, this variable will contain the alternate location string returned by the BMI server. In its default configuration, this is a header-like string that can be added to the message with "headers_add". If there is no verdict available for the recipient being routed, or if the message is to be delivered normally, this variable contains the empty string. o $bmi_deliver This is an additional integer variable that can be used to query if the message should be delivered at all. You should use router preconditions instead if possible. $bmi_deliver is '0': the message should NOT be delivered. $bmi_deliver is '1': the message should be delivered. IMPORTANT NOTE: Verdict inheritance. The message is passed to the BMI server during message reception, using the target addresses from the RCPT TO: commands in the SMTP transaction. If recipients get expanded or re-written (for example by aliasing), the new address(es) inherit the verdict from the original address. This means that verdicts also apply to all "child" addresses generated from top-level addresses that were sent to the BMI server. 5) Using per-recipient opt-in information (Optional) The BMI server features multiple scanning "profiles" for individual recipients. These are usually stored in a LDAP server and are queried by the BMI server itself. However, you can also pass opt-in data for each recipient from the MTA to the BMI server. This is particularly useful if you already look up recipient data in exim anyway (which can also be stored in a SQL database or other source). This implementation enables you to pass opt-in data to the BMI server in the RCPT ACL. This works by setting the 'bmi_optin' modifier in a block of that ACL. If should be set to a list of comma-separated strings that identify the features which the BMI server should use for that particular recipient. Ideally, you would use the 'bmi_optin' modifier in the same ACL block where you set the 'bmi_run' control flag. Here is an example that will pull opt-in data for each recipient from a flat file called '/etc/exim/bmi_optin_data'. The file format: user1@mydomain.com: : user2@thatdomain.com: The example: accept domains = +relay_to_domains endpass verify = recipient bmi_optin = ${lookup{$local_part@$domain}lsearch{/etc/exim/bmi_optin_data}} control = bmi_run Of course, you can also use any other lookup method that exim supports, including LDAP, Postgres, MySQL, Oracle etc., as long as the result is a list of colon-separated opt-in strings. For a list of available opt-in strings, please contact your Brightmail representative. 8. Sender Policy Framework (SPF) support -------------------------------------------------------------- To learn more about SPF, visit http://spf.pobox.com. This document does not explain the SPF fundamentals, you should read and understand the implications of deploying SPF on your system before doing so. SPF support is added via the libspf2 library. Visit http://www.libspf2.org/ to obtain a copy, then compile and install it. By default, this will put headers in /usr/local/include and the static library in /usr/local/lib. To compile exim with SPF support, set these additional flags in Local/Makefile: CFLAGS=-DSPF -I/usr/local/include EXTRALIBS_EXIM=-L/usr/local/lib -lspf2 This assumes that the libspf2 files are installed in their default locations. You can now run SPF checks in incoming SMTP by using the "spf" ACL condition in either the MAIL, RCPT or DATA ACLs. When using it in the RCPT ACL, you can make the checks dependend on the RCPT address (or domain), so you can check SPF records only for certain target domains. This gives you the possibility to opt-out certain customers that do not want their mail to be subject to SPF checking. The spf condition takes a list of strings on its right-hand side. These strings describe the outcome of the SPF check for which the spf condition should succeed. Valid strings are: o pass The SPF check passed, the sending host is positively verified by SPF. o fail The SPF check failed, the sending host is NOT allowed to send mail for the domain in the envelope-from address. o softfail The SPF check failed, but the queried domain can't absolutely confirm that this is a forgery. o none The queried domain does not publish SPF records. o neutral The SPF check returned a "neutral" state. This means the queried domain has published a SPF record, but wants to allow outside servers to send mail under its domain as well. o err_perm This indicates a syntax error in the SPF record of the queried domain. This should be treated like "none". o err_temp This indicates a temporary error during all processing, including exim's SPF processing. You may defer messages when this occurs. You can prefix each string with an exclamation mark to invert is meaning, for example "!fail" will match all results but "fail". The string list is evaluated left-to-right, in a short-circuit fashion. When a string matches the outcome of the SPF check, the condition succeeds. If none of the listed strings matches the outcome of the SPF check, the condition fails. Here is a simple example to fail forgery attempts from domains that publish SPF records: /* ----------------- deny message = $sender_host_address is not allowed to send mail from $sender_address_domain spf = fail --------------------- */ You can also give special treatment to specific domains: /* ----------------- deny message = AOL sender, but not from AOL-approved relay. sender_domains = aol.com spf = fail:neutral --------------------- */ Explanation: AOL publishes SPF records, but is liberal and still allows non-approved relays to send mail from aol.com. This will result in a "neutral" state, while mail from genuine AOL servers will result in "pass". The example above takes this into account and treats "neutral" like "fail", but only for aol.com. Please note that this violates the SPF draft. When the spf condition has run, it sets up several expansion variables. $spf_header_comment This contains a human-readable string describing the outcome of the SPF check. You can add it to a custom header or use it for logging purposes. $spf_received This contains a complete SPF-Received: header that can be added to the message. Please note that according to the SPF draft, this header must be added at the top of the header list. Please see section 10 on how you can do this. $spf_result This contains the outcome of the SPF check in string form, one of pass, fail, softfail, none, neutral, err_perm or err_temp. $spf_smtp_comment This contains a string that can be used in a SMTP response to the calling party. Useful for "fail". 9. SRS (Sender Rewriting Scheme) Support -------------------------------------------------------------- Exiscan currently includes SRS support via Miles Wilton's libsrs_alt library. His patch to link to that library is included in exiscan. The current version of the supported library is 0.5. In order to use SRS, you must get a copy of libsrs_alt from http://srs.mirtol.com/ Unpack the tarball, then refer to MTAs/README.EXIM to proceed. The SRS support is subject to change and should be considered experimental. SRS itself is still subject to changes. 10. Selecting header positions when adding headers in ACLs --------------------------------------------------------------\ Exim introduced selectable header positions in Version 4.43. One option allows you to add your header after all Received: headers. For optimal SPF draft/RFC2822 compliance, exiscan-acl adds one other option, called "spf-received". It adds your header before the first header which is NOT Received: or Resent-*:, seen from the top of the header stack. Example: warn message = :spf_received:$spf_received -------------------------------------------------------------- End of file --------------------------------------------------------------