NAME

postfwd3 - postfix firewall daemon

SYNOPSIS

postfwd3 [OPTIONS] [COMMAND]

postfwd3 [OPTIONS] --cmd [SOURCE1, SOURCE2, ...]

        Ruleset: (at least one, multiple use is allowed):
        -f, --file <file>               reads rules from <file>
        -r, --rule <rule>               adds <rule> to config
        -s, --scores <v>=<r>            returns <r> when score exceeds <v>

        Settings: (multiple use allowed)
        -F, --loadsettings <file>       loads program settings from <file>
            --savesettings <file>       saves program settings to <file>
            --showsettings              exports program settings to stdout

        Server:
        -i, --interface <dev>           listen on interface <dev>
        -p, --port <port>               listen on port <port>
            --proto <proto>             socket type (tcp or unix)
            --server_socket <sock>      e.g. tcp:127.0.0.1:10045
        -u, --user <name>               set uid to user <name>
        -g, --group <name>              set gid to group <name>
            --umask <mask>              umask for master filepermissions
            --server_umask <mask>       umask for server filepermissions
            --pidfile <path>            create pidfile under <path>
            --min_servers <i>           spawn at least <i> children
            --max_servers <i>           do not spawn more than <i> children
            --min_spare_servers <i>     minimum idle children
            --max_spare_servers <i>     maximum idle children

        Cache:
        -c, --cache <int>               sets the request-cache timeout to <int> seconds
            --cleanup-requests <int>    cleanup interval in seconds for request cache
            --cache_interface <dev>     listen on interface <dev>
            --cache_port <port>         listen on port <port>
            --cache_proto <proto>       socket type (tcp or unix)
            --cache_socket <sock>       e.g. tcp:127.0.0.1:10043
            --cache_umask <mask>        umask for cache filepermissions
            --cacheid <list>            list of request items for cache-id
            --cacheid_md5               cacheid => md5sum(request)
            --cache-rdomain-only        skip recipient localpart for cache-id
            --cache-no-sender           skip sender address for cache-id
            --cache-no-size             skip size for cache-id
            --no_parent_request_cache   disable parent request cache
            --no_parent_rate_cache      disable parent rate cache
            --no_parent_dns_cache       disable parent dns cache (default)
            --no_parent_group_cache     disable parent group cache
            --no_parent_cache           disable all parent caches

        Groups:
            --default_group_ttl <i>     default group TTL
            --group_maxitems <i>        max items per group
            --cleanup-groups <int>      cleanup interval in seconds for group objects

        Rates:
            --cleanup-rates <int>       cleanup interval in seconds for rate cache

        Control:
        -k, --kill, --stop              terminate postfwd3
            --reload, --hup             reload postfwd3
            --watchdog <w>              watchdog timer in seconds
            --respawn <r>               respawn delay in seconds
            --failures <f>              max respawn failure counter
        -d, --daemon                    execute program in background
            --nodaemon                  execute program in foreground
            --daemons <list>            list of daemons to start
            --personality <type>        type of policy server, allows 'PreFork' or 'Multiplex'
            --autopersonality           determine personality by program name (see manpage)
            --noautopersonality         don't do it (see above :)
            --v1                        set personality to 'Multiplex'
            --v2                        set personality to 'PreFork'
            --dumpcache                 show cache contents
            --dumpstats                 show statistics
        -R, --chroot <path>             chroot to <path> before start
            --delcache <item>           removes an item from the request cache
            --delrate <item>            removes an item from the rate cache

        DNS:
        -n, --nodns                     skip any dns based test
            --dns_timeout <i>           dns query timeout in seconds
            --dns_timeout_max <i>       disable dnsbl after <i> timeouts
            --dns_timeout_interval <i>  reenable dnsbl after <i> seconds
            --cache-rbl-timeout <i>     default dns ttl if not specified in ruleset
            --cache-rbl-default <s>     default dns pattern if not specified in ruleset
            --cleanup-rbls <i>          cleanup old dns cache items every <i> seconds
            --dns_async_txt             perform dnsbl A and TXT lookups simultaneously
            --dns_max_ns_lookups        max names to look up with sender_ns_addrs
            --dns_max_mx_lookups        max names to look up with sender_mx_addrs
            --ipv6_dnsbl                enables dnsbl checks for IPv6 addresses

        Optional:
        -t, --test                      testing, always returns "dunno"
        -S, --summary <i>               show stats every <i> seconds
            --noidlestats               disables statistics when idle
            --norulestats               disables per rule statistics
        -I, --instantcfg                reloads ruleset on every new request
            --config_timeout <i>        parser timeout in seconds
            --keep_groups               do not clear group cache on reload
            --save_groups <file>        save and load group cache on disk
            --keep_rates                do not clear rate limit counters on reload
            --save_rates <file>         save and load rate limits on disk
        -A, --aggregate_addrs           pre-compute ip address lists to subnets
            --no_netaddr                don't use NetAddr::IP functions
            --no_netcidr                don't use Net::CIDR::Lite functions
            --cidr_method=s             use method <s> for network checks


        Plugins:
            --plugins <file>            loads postfwd plugins from file

        Logging:
        -l, --logname <label>           label for syslog messages
            --facility <s>              use syslog facility <s>
            --socktype <s>              use syslog socktype <s>
            --nodnslog                  do not log dns results
            --anydnslog                 log any dns (even cached) results
            --norulelog                 do not log rule actions
            --nolog|--perfmon           no logging at all
        -v, --verbose                   verbose logging, use twice to increase
            --debug <s>                 list of debugging classes
            --debugclasses              shows all available debug classes
                                        and exits the program

        Information (use only at command-line!):
        -h, --help                      display this help and exit
        -m, --manual                    shows program manual
        -V, --version                   output version information and exit
        -D, --defaults                  show postfwd3 settings and exit
        -C, --showconfig                show postfwd3 ruleset and exit (-v allowed)
        -L, --stdout                    redirect syslog messages to stdout (--stdoutlog works for compatibility)
            --stdin                     pull request from stdin instead of a network socket
            --cmd                       shorthand form of --stdin, --stdout and --nodaemon
        -q, --quiet                     no syslogging, no stdout (-P works for compatibility)

        Commands:
        start                           start the program [default]
        stop                            same as --stop
        reload                          same as --reload
        dumpcache                       same as --dumpcache
        dumpstats                       same as --dumpstats
        defaults                        same as --defaults
        showsettings                    same as --showsettings
        delcache <item>                 same as --delcache <item>
        delrate <item>                  same as --delrate <item>

        Obsolete (only for compatibility with postfwd v1):
        --shortlog, --dns_queuesize, --dns_retries

DESCRIPTION

INTRODUCTION

postfwd3 is written to combine complex postfix restrictions in a ruleset similar to those of the most firewalls. The program uses the postfix policy delegation protocol to control access to the mail system before a message has been accepted (please visit http://www.postfix.org/SMTPD_POLICY_README.html for more information).

postfwd3 allows you to choose an action (e.g. reject, dunno) for a combination of several smtp parameters (like sender and recipient address, size or the client's TLS fingerprint). Also it offers simple macros/acls which should allow straightforward and easy-to-read configurations.

Features:

* Complex combinations of smtp parameters

* Combined RBL/RHSBL lookups with arbitrary actions depending on results

* Scoring system

* Date/time based rules

* Macros/ACLs, Dynamic groups, Negation

* Compare request attributes (e.g. client_name and helo_name)

* Internal caching for requests and dns lookups

* Built in statistics for rule efficiency analysis

CONFIGURATION

A configuration line consists of optional item=value pairs, separated by semicolons (`;`) and the appropriate desired action:

        [ <item1>=<value>; <item2>=<value>; ... ] action=<result>

Example:

        client_address=192.168.1.1 ; sender==no@bad.local ; action=REJECT

This will deny all mail from 192.168.1.1 with envelope sender no@bad.local. The order of the elements is not important. So the following would lead to the same result as the previous example:

        action=REJECT ; client_address=192.168.1.1 ; sender==no@bad.local

The way how request items are compared to the ruleset can be influenced in the following way:

        ====================================================================
         ITEM == VALUE                true if ITEM equals VALUE
         ITEM => VALUE                true if ITEM >= VALUE
         ITEM =< VALUE                true if ITEM <= VALUE
         ITEM >  VALUE                true if ITEM >  VALUE
         ITEM <  VALUE                true if ITEM <  VALUE
         ITEM =~ VALUE                true if ITEM ~= /^VALUE$/i
         ITEM != VALUE                false if ITEM equals VALUE
         ITEM !> VALUE                false if ITEM >= VALUE
         ITEM !< VALUE                false if ITEM <= VALUE
         ITEM !~ VALUE                false if ITEM ~= /^VALUE$/i
         ITEM =  VALUE                default behaviour (see ITEMS section)
        ====================================================================

To identify single rules in your log files, you may add an unique identifier for each of it:

        id=R_001 ; action=REJECT ; client_address=192.168.1.1 ; sender==no@bad.local

You may use these identifiers as target for the `jump()` command (see ACTIONS section below). Leading or trailing whitespace characters will be ignored. Use '#' to comment your configuration. Others will appreciate.

A ruleset consists of one or multiple rules, which can be loaded from files or passed as command line arguments. Please see the COMMAND LINE section below for more information on this topic.

Since postfwd version 1.30 rules spanning span multiple lines can be defined by prefixing the following lines with one or multiple whitespace characters (or '}' for macros):

        id=RULE001
                client_address=192.168.1.0/24
                sender==no@bad.local
                action=REJECT no access

postfwd versions prior to 1.30 require trailing ';' and '\'-characters:

        id=RULE001; \
                client_address=192.168.1.0/24; \
                sender==no@bad.local; \
                action=REJECT no access

ITEMS

        id                      - a unique rule id, which can be used for log analysis
                                  ids also serve as targets for the "jump" command.

        date, time              - a time or date range within the specified rule shall hit
                                  # FORMAT:
                                  # Feb, 29th
                                  date=29.02.2008
                                  # Dec, 24th - 26th
                                  date=24.12.2008-26.12.2008
                                  # from today until Nov, 23rd
                                  date=-23.09.2008
                                  # from April, 1st until today
                                  date=01.04.2008-

        days, months            - a range of weekdays (Sun-Sat) or months (Jan-Dec)
                                  within the specified rule shall hit

        score                   - when the specified score is hit (see ACTIONS section)
                                  the specified action will be returned to postfix
                                  scores are set global until redefined!

        request_score           - this value allows to access a request's score. it
                                  may be used as variable ($$request_score).

        rbl, rhsbl,             - query the specified RBLs/RHSBLs, possible values are:
        rhsbl_client,             <name>[/<reply>/<maxcache>, <name>/<reply>/<maxcache>]
        rhsbl_sender,             (defaults: reply=^127\.0\.0\.\d+$ maxcache=3600)
        rhsbl_reverse_client      the results of all rhsbl_* queries will be combined
                                  in rhsbl_count (see below).

        rblcount, rhsblcount    - minimum RBL/RHSBL hitcounts to match. if not specified
                                  a single RBL/RHSBL hit will match the rbl/rhsbl items.
                                  you may specify 'all' to evaluate all items, and use
                                  it as variable in an action (see ACTIONS section)
                                  (default: 1)

        sender_localpart,       - the local-/domainpart of the sender address
        sender_domain

        recipient_localpart,    - the local-/domainpart of the recipient address
        recipient_domain

        helo_address            - postfwd3 tries to look up the helo_name. use
                                  helo_address=!!(0.0.0.0/0) to check for unknown.
                                  Please do not use this for positive access control
                                  (whitelisting), as it might be forged.

        sender_ns_names,        - postfwd3 tries to look up the names/ip addresses
        sender_ns_addrs           of the nameservers for the sender domain part.
                                  Please do not use this for positive access control
                                  (whitelisting), as it might be forged.

        sender_mx_names,        - postfwd3 tries to look up the names/ip addresses
        sender_mx_addrs           of the mx records for the sender domain part.
                                  Please do not use this for positive access control
                                  (whitelisting), as it might be forged.

        version                 - postfwd3 version, contains "postfwd3 n.nn"
                                  this enables version based checks in your rulesets
                                  (e.g. for migration). works with old versions too,
                                  because a non-existing item always returns false:
                                  # version >= 1.10
                                  id=R01; version~=1\.[1-9][0-9]; sender_domain==some.org \
                                        ; action=REJECT sorry no access

        postfwd_port            - postfwd server port, allows to use the same ruleset different
                                  instances:
                                  # rule only hits for instance at port 10045
                                  id=PORT10045; postfwd_port==10045; action=DUNNO

        postfwd_interface       - postfwd server initerface, see postfwd_port

        ratecount               - only available for rate(), size() and rcpt() actions.
                                  contains the actual limit counter:
                                        id=R01; action=rate(sender/200/600/REJECT limit of 200 exceeded [$$ratecount hits])
                                        id=R02; action=rate(sender/100/600/WARN limit of 100 exceeded [$$ratecount hits])

Besides these you can specify any attribute of the postfix policy delegation protocol. Feel free to combine them the way you need it (have a look at the EXAMPLES section below).

Most values can be specified as regular expressions (PCRE). Please see the table below for details:

        # ==========================================================
        # ITEM=VALUE                            TYPE
        # ==========================================================
        id=something                            mask = string
        date=01.04.2007-22.04.2007              mask = date (DD.MM.YYYY-DD.MM.YYYY)
        time=08:30:00-17:00:00                  mask = time (HH:MM:SS-HH:MM:SS)
        days=Mon-Wed                            mask = weekdays (Mon-Wed) or numeric (1-3)
        months=Feb-Apr                          mask = months (Feb-Apr) or numeric (1-3)
        score=5.0                               mask = maximum floating point value
        rbl=zen.spamhaus.org                    mask = <name>/<reply>/<maxcache>[,...]
        rblcount=2                              mask = numeric, will match if rbl hits >= 2
        helo_address=<a.b.c.d/nn>               mask = CIDR[,CIDR,...]
        sender_ns_names=some.domain.tld         mask = PCRE
        sender_mx_names=some.domain.tld         mask = PCRE
        sender_ns_addrs=<a.b.c.d/nn>            mask = CIDR[,CIDR,...]
        sender_mx_addrs=<a.b.c.d/nn>            mask = CIDR[,CIDR,...]
        # ------------------------------
        # Postfix version 2.1 and later:
        # ------------------------------
        client_address=<a.b.c.d/nn>             mask = CIDR[,CIDR,...]
        client_name=another.domain.tld          mask = PCRE
        reverse_client_name=another.domain.tld  mask = PCRE
        helo_name=some.domain.tld               mask = PCRE
        sender=foo@bar.tld                      mask = PCRE
        recipient=bar@foo.tld                   mask = PCRE
        recipient_count=5                       mask = numeric, will match if recipients >= 5
        # ------------------------------
        # Postfix version 2.2 and later:
        # ------------------------------
        sasl_method=plain                       mask = PCRE
        sasl_username=you                       mask = PCRE
        sasl_sender=                            mask = PCRE
        size=12345                              mask = numeric, will match if size >= 12345
        ccert_subject=blackhole.nowhere.local   mask = PCRE (only if tls verified)
        ccert_issuer=John+20Doe                 mask = PCRE (only if tls verified)
        ccert_fingerprint=AA:BB:CC:DD:EE:...    mask = PCRE (do NOT use "..." here)
        # ------------------------------
        # Postfix version 2.3 and later:
        # ------------------------------
        encryption_protocol=TLSv1/SSLv3         mask = PCRE
        encryption_cipher=DHE-RSA-AES256-SHA    mask = PCRE
        encryption_keysize=256                  mask = numeric, will match if keysize >= 256
        ...

the current list can be found at http://www.postfix.org/SMTPD_POLICY_README.html. Please read carefully about which attribute can be used at which level of the smtp transaction (e.g. size will only work reliably at END-OF-MESSAGE level). Pattern matching is performed case insensitive.

Multiple use of the same item is allowed and will compared as logical OR, which means that this will work as expected:

        id=TRUST001; action=OK; encryption_keysize=64
                ccert_fingerprint=11:22:33:44:55:66:77:88:99
                ccert_fingerprint=22:33:44:55:66:77:88:99:00
                ccert_fingerprint=33:44:55:66:77:88:99:00:11
                sender=@domain\.local$

client_address, rbl and rhsbl items may also be specified as whitespace-or-comma-separated values:

        id=SKIP01; action=dunno
                client_address=192.168.1.0/24, 172.16.254.23
        id=SKIP02; action=dunno
                client_address= 10.10.3.32 10.216.222.0/27

The following items must be unique:

        id, minimum and maximum values, rblcount and rhsblcount

Any item can be negated by preceeding '!!' to it, e.g.:

        id=HOST001 ;  hostname == !!secure.trust.local ;  action=REJECT only secure.trust.local please

or using the right compare operator:

        id=HOST001 ;  hostname != secure.trust.local ;  action=REJECT only secure.trust.local please

To avoid confusion with regexps or simply for better visibility you can use '!!(...)':

        id=USER01 ;  sasl_username =~ !!( /^(bob|alice)$/ )  ;  action=REJECT who is that?

Request attributes can be compared by preceeding '$$' characters, e.g.:

        id=R-003 ;  client_name = !! $$helo_name      ;  action=WARN helo does not match DNS
        # or
        id=R-003 ;  client_name = !!($$(helo_name))   ;  action=WARN helo does not match DNS

This is only valid for PCRE values (see list above). The comparison will be performed as case insensitive exact match. Use the '-vv' option to debug.

These special items will be reset for any new rule:

        rblcount        - contains the number of RBL answers
        rhsblcount      - contains the number of RHSBL answers
        matches         - contains the number of matched items
        dnsbltext       - contains the dns TXT part of all RBL and RHSBL replies in the form
                          rbltype:rblname:<txt>; rbltype:rblname:<txt>; ...

These special items will be changed for any matching rule:

        request_hits    - contains ids of all matching rules

This means that it might be necessary to save them, if you plan to use these values in later rules:

        # set vals
        id=RBL01 ; rhsblcount=all; rblcount=all
                action=set(HIT_rhls=$$rhsblcount,HIT_rbls=$$rblcount,HIT_txt=$$dnsbltext)
                rbl=list.dsbl.org, bl.spamcop.net, dnsbl.sorbs.net, zen.spamhaus.org
                rhsbl_client=rddn.dnsbl.net.au, rhsbl.ahbl.org, rhsbl.sorbs.net
                rhsbl_sender=rddn.dnsbl.net.au, rhsbl.ahbl.org, rhsbl.sorbs.net

        # compare
        id=RBL02 ; HIT_rhls>=1 ; HIT_rbls>=1 ; action=554 5.7.1 blocked using $$HIT_rhls RHSBLs and $$HIT_rbls RBLs [INFO: $$HIT_txt]
        id=RBL03 ; HIT_rhls>=2               ; action=554 5.7.1 blocked using $$HIT_rhls RHSBLs [INFO: $$HIT_txt]
        id=RBL04 ; HIT_rbls>=2               ; action=554 5.7.1 blocked using $$HIT_rbls RBLs [INFO: $$HIT_txt]

FILES

Since postfwd1 v1.15 and postfwd3 v0.18 long item lists can be stored in separate files:

        id=R001 ;  ccert_fingerprint==file:/etc/postfwd/wl_ccerts ;  action=DUNNO

postfwd3 will read a list of items (one item per line) from /etc/postfwd/wl_ccerts. comments are allowed:

        # client1
        11:22:33:44:55:66:77:88:99
        # client2
        22:33:44:55:66:77:88:99:00
        # client3
        33:44:55:66:77:88:99:00:11

To use existing tables in key=value format, you can use:

        id=R001 ;  ccert_fingerprint==table:/etc/postfwd/wl_ccerts ;  action=DUNNO

This will ignore the right-hand value. Items can be mixed:

        id=R002 ;  action=REJECT
                client_name==unknown
                client_name==file:/etc/postfwd/blacklisted

and for non pcre (comma separated) items:

        id=R003 ;  action=REJECT
                client_address==10.1.1.1, file:/etc/postfwd/blacklisted

        id=R004 ;  action=REJECT
                rbl=myrbl.home.local, zen.spamhaus.org, file:/etc/postfwd/rbls_changing

You can check your configuration with the --show_config option at the command line:

        # postfwd3 --showconfig --rule='action=DUNNO; client_address=10.1.0.0/16, file:/etc/postfwd/wl_clients, 192.168.2.1'

should give something like:

        Rule   0: id->"R-0"; action->"DUNNO"; client_address->"=;10.1.0.0/16, =;194.123.86.10, =;186.4.6.12, =;192.168.2.1"

If a file can not be read, it will be ignored:

        # postfwd3 --showconfig --rule='action=DUNNO; client_address=10.1.0.0/16, file:/etc/postfwd/wl_clients, 192.168.2.1'
        [LOG warning]: error: file /etc/postfwd/wl_clients not found - file will be ignored ?
        Rule   0: id->"R-0"; action->"DUNNO"; client_address->"=;10.1.0.0/16, =;192.168.2.1"

File items are evaluated at configuration stage. Therefore postfwd3 needs to be reloaded if a file has changed

If you want to specify a file, that will be reloaded for each request, you can use lfile: and ltable:

        id=R001; client_address=lfile:/etc/postfwd/client_whitelist; action=dunno

This will check the modification time of /etc/postfwd/client_whitelist every time the rule is evaluated and reload it as necessary. Of course this might increase the system load, so please use it with care.

The --showconfig option illustrates the difference:

        ## evaluated at configuration stage
        # postfwd3 --nodaemon -L --rule='client_address=table:/etc/postfwd/clients; action=dunno' -C
        Rule   0: id->"R-0"; action->"dunno"; client_address->"=;1.1.1.1, =;1.1.1.2, =;1.1.1.3"

        ## evaluated for any rulehit
        # postfwd3 --nodaemon -L --rule='client_address=ltable:/etc/postfwd/clients; action=dunno' -C
        Rule   0: id->"R-0"; action->"dunno"; client_address->"=;ltable:/etc/postfwd/clients"

Files can refer to other files. The following is valid.

        -- FILE /etc/postfwd/rules.cf --
        id=R01; client_address=file:/etc/postfwd/clients_master.cf; action=DUNNO

        -- FILE /etc/postfwd/clients_master.cf --
        192.168.1.0/24
        file:/etc/postfwd/clients_east.cf
        file:/etc/postfwd/clients_west.cf

        -- FILE /etc/postfwd/clients_east.cf --
        192.168.2.0/24

        -- FILE /etc/postfwd/clients_west.cf --
        192.168.3.0/24

Note that there is currently no loop detection (/a/file calls /a/file) and that this feature is only available with postfwd1 v1.15 and postfwd3 v0.18 and higher.

ACTIONS

General

Actions will be executed, when all rule items have matched a request (or at least one of any item list). You can refer to request attributes by preceeding $$ characters, like:

        id=R-003; client_name = !!$$helo_name; action=WARN helo '$$helo_name' does not match DNS '$$client_name'
        # or
        id=R-003; client_name = !!$$helo_name; action=WARN helo '$$(helo_name)' does not match DNS '$$(client_name)'

postfix actions

Actions will be replied to postfix as result to policy delegation requests. Any action that postfix understands is allowed - see "man 5 access" or http://www.postfix.org/access.5.html for a description. If no action is specified, the postfix WARN action which simply logs the event will be used for the corresponding rule.

postfwd3 will return dunno if it has reached the end of the ruleset and no rule has matched. This can be changed by placing a last rule containing only an action statement:

        ...
        action=dunno ; sender=@domain.local     # sender is ok
        action=reject                           # default deny

postfwd3 actions

postfwd3 actions control the behaviour of the program. Currently you can specify the following:

        jump (<id>)
        jumps to rule with id <id>, use this to skip certain rules.
        you can jump backwards - but remember that there is no loop
        detection at the moment! jumps to non-existing ids will be skipped.

        score (<score>)
        the request's score will be modified by the specified <score>,
        which must be a floating point value. the modificator can be either
                +n.nn   adds n.nn to current score
                -n.nn   sustracts n.nn from the current score
                *n.nn   multiplies the current score by n.nn
                /n.nn   divides the current score through n.nn
                =n.nn   sets the current score to n.nn
        if the score exceeds the maximum set by `--scores` option (see
        COMMAND LINE) or the score item (see ITEMS section), the action
        defined for this case will be returned (default: 5.0=>"REJECT postfwd3 score exceeded").

        set (<item>=<value>,<item>=<value>,...)
        this command allows you to insert or override request attributes, which then may be
        compared to your further ruleset. use this to speed up repeated comparisons to large item lists.
        please see the EXAMPLES section for more information. you may separate multiple key=value pairs
        by "," characters.

        rate (<item>/<max>/<time>/<action>)
        this command creates a counter for the given <item>, which will be increased any time a request
        containing it arrives. if it exceeds <max> within <time> seconds it will return <action> to postfix.
        rate counters are very fast as they are executed before the ruleset is parsed.
        please note that <action> was limited to postfix actions (no postfwd actions) for postfwd versions <1.33!
            # no more than 3 requests per 5 minutes
            # from the same "unknown" client
            id=RATE01 ;  client_name==unknown
               action=rate(client_address/3/300/450 4.7.1 sorry, max 3 requests per 5 minutes)

        size (<item>/<max>/<time>/<action>)
        this command works similar to the rate() command with the difference, that the rate counter is
        increased by the request's size attribute. to do this reliably you should call postfwd3 from
        smtpd_end_of_data_restrictions. if you want to be sure, you could check it within the ruleset:
           # size limit 1.5mb per hour per client
           id=SIZE01 ;  protocol_state==END-OF-MESSAGE ;  client_address==!!(10.1.1.1)
              action=size(client_address/1572864/3600/450 4.7.1 sorry, max 1.5mb per hour)

        rcpt (<item>/<max>/<time>/<action>)
        this command works similar to the rate() command with the difference, that the rate counter is
        increased by the request's recipient_count attribute. to do this reliably you should call postfwd
        from smtpd_data_restrictions or smtpd_end_of_data_restrictions. if you want to be sure, you could
        check it within the ruleset:
           # recipient count limit 3 per hour per client
           id=RCPT01 ;  protocol_state==END-OF-MESSAGE ;  client_address==!!(10.1.1.1)
              action=rcpt(client_address/3/3600/450 4.7.1 sorry, max 3 recipients per hour)

        rate5321,size5321,rcpt5321 (<item>/<max>/<time>/<action>)
        same as the corresponding non-5321 functions, with the difference that the localpart of
        sender oder recipient addresses are evaluated case-sensitive according to rfc5321. That
        means that requests from bob@example.local and BoB@example.local will be treated differently

        groupadd(<groupname>/<item>[/<ttl>])
        saves <item> to dynamic group <group> for later use. the object will be removed from that
        group after ttl seconds. if not set, postfwd3 will use the default --default_group_ttl [3600s].
        Please read the chapter about dynamic groups before use!
           # add client_address to group %%blacklisted_hosts
           id=ADDGROUP
                rbl=zen.spamhaus.org,bl.spamcop.net,ix.dnsbl.manitu.net
                action=groupadd(%%blacklisted_hosts/client_address/86400)

        groupdel(<groupname>/<item>)
        removes <item> from dynamic group <group>. Please read the chapter about dynamic groups before use!
           # remove a client_address from group %%blacklisted_hosts
           id=DELGROUP
                rbl=lists.dnswl.org
                action=groupdel(%%blacklisted_hosts/client_address)

        ask (<addr>:<port>[:<ignore>])
        allows to delegate the policy decision to another policy service (e.g. postgrey). the first
        and the second argument (address and port) are mandatory. a third optional argument may be
        specified to tell postfwd3 to ignore certain answers and go on parsing the ruleset:
           # example1: query postgrey and return it's answer to postfix
           id=GREY; client_address==10.1.1.1; action=ask(127.0.0.1:10031)
           # example2: query postgrey but ignore it's answer, if it matches 'DUNNO'
           # and continue parsing postfwd's ruleset
           id=GREY; client_address==10.1.1.1; action=ask(127.0.0.1:10031:^dunno$)

        mail(server/helo/from/to/subject/body)
        This command is deprecated. You should try to use the sendmail() action instead.
        Very basic mail command, that sends a message with the given arguments. LIMITATIONS:
        This basically performs a telnet. No authentication or TLS are available. Additionally it does
        not track notification state and will notify you any time, the corresponding rule hits.

        sendmail(sendmail-path::from::to::subject::body)
        Mail command, that uses an existing sendmail binary and sends a message with the given arguments.
        LIMITATIONS: The command does not track notification state and will notify you any time, the
        corresponding rule hits (which could mean 100 mails for a mail with 100 recipients at RCPT stage).

        wait (<delay>)
        pauses the program execution for <delay> seconds. use this for
        delaying or throtteling connections.

        note (<string>)
        just logs the given string and continues parsing the ruleset.
        if the string is empty, nothing will be logged (noop).

        quit (<code>)
        terminates the program with the given exit-code. postfix doesn`t
        like that too much, so use it with care.

You can reference to request attributes, like

        id=R-HELO ;  helo_name=^[^\.]+$ ;  action=REJECT invalid helo '$$helo_name'

Since postfwd3 version 2.00 a rule can have multiple postfwd actions, like

        # if client is found on list.dnswl.org,
        #   1. send a note
        #   2. add it's ip to group %%WHITELISTED
        #   3. permit the request
        id=ADDR01
                rbl=list.dnswl.org
                action=note(adding $$client_address to WHITELIST)
                action=groupadd(%%WHITELISTED/client_address)
                action=PERMIT_AUTH_DESTINATION

MACROS/ACLS

Multiple use of long items or combinations of them may be abbreviated by macros. Those must be prefixed by '&&' (two '&' characters). First the macros have to be defined as follows:

        &&RBLS { rbl=zen.spamhaus.org,list.dsbl.org,bl.spamcop.net,dnsbl.sorbs.net,ix.dnsbl.manitu.net; };

Then these may be used in your rules, like:

        &&RBLS ;  client_name=^unknown$                         ; action=REJECT
        &&RBLS ;  client_name=(\d+[\.-_]){4}                    ; action=REJECT
        &&RBLS ;  client_name=[\.-_](adsl|dynamic|ppp|)[\.-_]   ; action=REJECT

Macros can contain actions, too:

        # definition
        &&GONOW { action=REJECT your request caused our spam detection policy to reject this message. More info at http://www.domain.local; };
        # rules
        &&GONOW ;  &&RBLS ;  client_name=^unknown$
        &&GONOW ;  &&RBLS ;  client_name=(\d+[\.-_]){4}
        &&GONOW ;  &&RBLS ;  client_name=[\.-_](adsl|dynamic|ppp|)[\.-_]

Macros can contain macros, too:

        # definition
        &&RBLS{
                rbl=zen.spamhaus.org
                rbl=list.dsbl.org
                rbl=bl.spamcop.net
                rbl=dnsbl.sorbs.net
                rbl=ix.dnsbl.manitu.net
        };
        &&DYNAMIC{
                client_name=^unknown$
                client_name=(\d+[\.-_]){4}
                client_name=[\.-_](adsl|dynamic|ppp|)[\.-_]
        };
        &&GOAWAY { &&RBLS; &&DYNAMIC; };
        # rules
        &&GOAWAY ; action=REJECT dynamic client and listed on RBL

Basically macros are simple text substitutions - see the "PARSER" section for more information.

DYNAMIC GROUPS

Dynamic groups are list objects. You can add or remove items and compare request items to their content within your ruleset. Groups are saved persistant for further requests. This way you can save information for later use. Every item of a group has a time-to-live value.

Group-names have to be preceeded by double '%'-characters, like %%GROUPNAME:

        # add client_address to group %%BLACKLISTED, if sent to spamtrap
        id=SPAMTRAP; recipient==spamtrap@domain.local; action=groupadd(%%BLACKLISTED/client_address)

        # reject any request from client_address in group %%BLACKLISTED
        id=BLACKLISTED; client_address=%%BLACKLISTED; action=REJECT

The groupadd() function receives the following arguments. The <ttl> value is optional. If not specified, postfwd3 will use the default --default_group_ttl [300s]:

        groupadd ( <Groupname> / <item> [ / <ttl> ] )

postfwd will save this data in the following structure:

        %Group_Cache -> %<Groupname> -> $<item> = <ttl>

The groupdel() function removes an item from a group:

        groupdel ( <Groupname> / <item> )

Expired members will be removed at cleanup stage after --cleanup_groups <i> seconds.

To compare requests with dynamic groups, these must be refered by preceeding '%%'-characters:

        <item> = %%<Groupname>

Empty groups always compare false. The following ruleset is possible:

        # allow any request from client_address in group %%WHITELISTED
        id=WHITEGROUP; client_address=%%WHITELISTED; action=DUNNO

        # reject any request from client_address in group %%BLACKLISTED
        id=BLACKGROUP; client_address=%%BLACKLISTED; action=REJECT

        # ... other rules ...

        # ... whitelist file ...
        id=W_TRUST; action=set(WHITE=1)
            client_address=file:/etc/postfwd/whitelist_networks

        # ... whitelist dns ...
        id=W_DNSWL; action=set(WHITE=1)
            rbl=list.dnswl.org

        # ... -> add it to whitelist for a day and allow request
        id=W_GROUP; WHITE==1
            action=groupadd(%%WHITELISTED/client_address/86400)
            action=DUNNO

        # ... or blacklist by sending to spamtrap ...
        id=B_SPAMTRAP; action=set(BLACK=1)
            recipient==spamtrap@domain.local

        # ... or blacklisted by dns ...
        id=B_DNSBL; action=set(BLACK=1)
            rbl=zen.spamhaus.org

        # ... -> add it to group BLACKLISTED for 1 hour and reject
        id=B_GROUP; BLACK==1
            action=groupadd(%%BLACKLISTED/client_address/3600)
            action=REJECT

Limitations

Currently dynamic groups are kept in memory. By default postfwd3 will only accept a maximum of 999999 members in a group. To deactivate this limit set --group_maxitems to '-1'.

With PreFork personality, you should be aware that this information must be shared through the cache daemon.

In short: Don't let these lists get excessively big unless you have enough system capacity to do so. If necessary tune your config by setting the <ttl> and --group_maxitems.

PLUGINS

Description

The plugin interface allow you to define your own checks and enhance postfwd's functionality. Feel free to share useful things!

Warning

Note that the plugin interface is still at devel stage. Please test your plugins carefully, because errors may cause postfwd to break! It is also allowed to override attributes or built-in functions, but be sure that you know what you do because some of them are used internally.

Please keep security in mind, when you access sensible ressources and never, ever run postfwd as privileged user! Also never trust your input (especially hostnames, and e-mail addresses).

ITEMS

Item plugins are perl subroutines which integrate additional attributes to requests before they are evaluated against postfwd's ruleset like any other item of the policy delegation protocol. This allows you to create your own checks.

plugin-items can not be used selective. these functions will be executed for every request postfwd receives, so keep performance in mind.

        SYNOPSIS: postfwd_items_plugin{<name>}($request)

means that your subroutine, called <name>, has access to a hash-reference called $request, which contains all request attributes, like $request->{client_name} and saves values in the following form:

        save: $result->{<item>} = <value>

this creates the new item <item> containing <value>, which will be integrated in the policy delegation request and therefore may be used in postfwd's ruleset.

        # do NOT remove the next line
        %postfwd_items_plugin = (

                # EXAMPLES - integrated in postfwd. no need to activate them here.
                
                        # allows to check postfwd version in ruleset
                        "version" => sub {
                                my($request) = shift;
                                $request->{version} => $NAME." ".$VERSION,
                        },
                
                        # sender_domain and recipient_domain
                        "address_parts" => sub {
                                my($request) = shift;
                                $request->{sender} =~ /@([^@]*)$/;
                                $request->{sender_domain} = ($1 || '');
                                $request->{recipient} =~ /@([^@]*)$/;
                                $request->{recipient_domain} = ($1 || '');
                        },

        # do NOT remove the next line
        );

COMPARE

Compare plugins allow you to define how your new items should be compared to the ruleset. These are optional. If you don't specify one, the default (== for exact match, =~ for PCRE, ...) will be used.

        SYNOPSIS:  <item> => sub { return &{$postfwd_compare{<type>}}(@_); },

        # do NOT remove the next line
        %postfwd_compare_plugin = (

                EXAMPLES - integrated in postfwd. no need to activate them here.
        
                        # Simple example
                        # SYNOPSIS:  <result> = <item> (return &{$postfwd_compare{<type>}}(@_))
                        "client_address"  => sub { return &{$postfwd_compare{cidr}}(@_); },
                        "size"            => sub { return &{$postfwd_compare{numeric}}(@_); },
                        "recipient_count" => sub { return &{$postfwd_compare{numeric}}(@_); },
        
                        # Complex example
                        # SYNOPSIS:  <result> = <item>(<operator>, <ruleset value>, <request value>, <request>)
                        "numeric" => sub {
                                my($cmp,$val,$myitem,$request) = @_;
                                my($myresult) = undef;  $myitem ||= "0"; $val ||= "0";
                                if ($cmp eq '==') {
                                        $myresult = ($myitem == $val);
                                } elsif ($cmp eq '=<') {
                                        $myresult = ($myitem <= $val);
                                } elsif ($cmp eq '=>') {
                                        $myresult = ($myitem >= $val);
                                } elsif ($cmp eq '<') {
                                        $myresult = ($myitem < $val);
                                } elsif ($cmp eq '>') {
                                        $myresult = ($myitem > $val);
                                } elsif ($cmp eq '!=') {
                                        $myresult = not($myitem == $val);
                                } elsif ($cmp eq '!<') {
                                        $myresult = not($myitem <= $val);
                                } elsif ($cmp eq '!>') {
                                        $myresult = not($myitem >= $val);
                                } else {
                                        $myresult = ($myitem >= $val);
                                };
                                return $myresult;
                        },

        # do NOT remove the next line
        );

ACTIONS

Action plugins allow to define new postfwd actions. By setting the $stop-flag you can decide to continue or to stop parsing the ruleset.

        SYNOPSIS:  (<stop rule parsing>, <next rule index>, <return action>, <logprefix>) =
                        <action> (<current rule index>, <current time>, <command name>, <argument>, <logprefix>, <request>)

        # do NOT remove the next line
        %postfwd_actions_plugin = (

                # EXAMPLES - integrated in postfwd. no need to activate them here.
        
                        # note(<logstring>) command
                        "note"  => sub {
                                my($index,$now,$mycmd,$myarg,$myline,$request) = @_;
                                my($myaction) = 'dunno'; my($stop) = 0;
                                log_info "[RULES] ".$myline." - note: ".$myarg if $myarg;
                                return ($stop,$index,$myaction,$myline);
                        },
        
                        # skips next <myarg> rules
                        "skip" => sub {
                                my($index,$now,$mycmd,$myarg,$myline,$request) = @_;
                                my($myaction) = 'dunno'; my($stop) = 0;
                                $index += $myarg if ( $myarg and not(($index + $myarg) > $#Rules) );
                                return ($stop,$index,$myaction,$myline);
                        },
        
                        # dumps current request contents to syslog
                        "dumprequest" => sub {
                                my($index,$now,$mycmd,$myarg,$myline,$request) = @_;
                                my($myaction) = 'dunno'; my($stop) = 0;
                                map { log_info "[DUMP] rule=$index, Attribute: $_=$request{$_}" } (keys %{$request});
                                return ($stop,$index,$myaction,$myline);
                        },

        # do NOT remove the next line
        );

COMMAND LINE

Ruleset

The following arguments are used to specify the source of the postfwd3 ruleset. This means that at least one of the following is required for postfwd3 to work.

        -f, --file <file>
        Reads rules from <file>. Please see the CONFIGURATION section
        for more information.

        -r, --rule <rule>
        Adds <rule> to ruleset. Remember that you might have to quote
        strings that contain whitespaces or shell characters.

Settings

        -F, --loadsettings <file>
        Loads program settings from <file>. data must be in Data::Dumper format.
        Please read the SETTINGS section for more information.

        --savesettings <file>
        Saves program settings to <file> in Data::Dumper format.
        Please read the SETTINGS section for more information.

        --showsettings
        Exports program settings in Data::Dumper format to stdout. This output
        can be saved to a file and later used with the --loadsettings option.

Scoring

        -s, --scores <val>=<action>
        Returns <action> to postfix, when the request's score exceeds <val>

Multiple usage is allowed. Just chain your arguments, like:

        postfwd3 -r "<item>=<value>;action=<result>" -f <file> -f <file> ...
          or
        postfwd3 --scores 4.5="WARN high score" --scores 5.0="REJECT postfwd3 score too high" ...

In case of multiple scores, the highest match will count. The order of the arguments will be reflected in the postfwd3 ruleset.

Networking

postfwd3 can be run as daemon so that it listens on the network for incoming requests. The following arguments will control it's behaviour in this case.

        -d, --daemon, --nodaemon
        postfwd3 will run as daemon and listen on the network for incoming
        queries (default 127.0.0.1:10045). Use --nodaemon to keep postfwd3
        running in foreground.

        -i, --interface <dev>
        Bind postfwd3 to the specified interface (default 127.0.0.1).

        -p, --port <port>
        postfwd3 listens on the specified port (default tcp/10045).

        --proto <type>
        The protocol type for postfwd's socket. Currently you may use 'tcp' or 'unix' here.
        To use postfwd3 with a unix domain socket, run it as follows:
            postfwd3 --proto=unix --port=/somewhere/postfwd.socket

        -u, --user <name>
        Changes real and effective user to <name>.

        -g, --group <name>
        Changes real and effective group to <name>.

        --personality <type>
        Type of policy server, allows 'PreFork' or 'Multiplex'.
        This option overrides --autopersonality.

        --v1
        set personality to 'Multiplex'
        This option overrides --autopersonality.

        --v2
        set personality to 'PreFork'
        This option overrides --autopersonality.

        --autopersonality
        postfwd3 determines the  personality by program name:
           * 'Multiplex' for 'postfwd' or 'postfwd1'
           * 'PreFork' for 'postfwd2'
        you can disable this by using --noautopersonality or
        explicitly using --personality, --v1 or --v2

        --umask <mask>
        Changes the umask for filepermissions of the master process (pidfile).
        Attention: This is umask, not chmod - you have to specify the bits that
        should NOT apply. E.g.: umask 077 equals to chmod 700.

        --cache_umask <mask>
        Changes the umask for filepermissions of the cache process (unix domain socket).

        --server_umask <mask>
        Changes the umask for filepermissions of the server process (unix domain socket).

        -R, --chroot <path>
        Chroot the process to the specified path.
        Please look at http://postfwd.org/postfwd3-chroot.html before use!

        --pidfile <path>
        The process id will be saved in the specified file.

        --facility <f>
        sets the syslog facility, default is 'mail'

        --socktype <s>
        sets the Sys::Syslog socktype to 'native', 'inet' or 'unix'.
        Default is to auto-detect this depening on module version and os.

        -l, --logname <label>
        Labels the syslog messages. Useful when running multiple
        instances of postfwd.

        --loglen <int>
        Truncates any syslog message after <int> characters.

Plugins

        --plugins <file>
        Loads postfwd plugins from file. Please see http://postfwd.org/postfwd.plugins
        or the plugins.postfwd.sample that is available from the tarball for more info.

Optional arguments

These parameters influence the way postfwd3 is working. Any of them can be combined.

        -v, --verbose
        Verbose logging displays a lot of useful information but can cause
        your logfiles to grow noticeably. So use it with caution. Set the option
        twice (-vv) to get more information (logs all request attributes).

        -c, --cache <int>    (default=600)
        Timeout for request cache, results for identical requests will be
        cached until config is reloaded or this time (in seconds) expired.
        A setting of 0 disables this feature.

        --cache-no-size
        Ignores size attribute for cache comparisons which will lead to better
        cache-hit rates. You should set this option, if you don't use the size
        item in your ruleset.

        --cache-no-sender
        Ignores sender address for cache comparisons which will lead to better
        cache-hit rates. You should set this option, if you don't use the sender
        item in your ruleset.

        --cache-rdomain-only 
        This will strip the localpart of the recipient's address before filling the
        cache. This may considerably increase cache-hit rates.

        --cache-rbl-timeout <timeout>     (default=3600)
        This default value will be used as timeout in seconds for rbl cache items,
        if not specified in the ruleset.

        --cache-rbl-default <pattern>    (default=^127\.0\.0\.\d+$)
        Matches <pattern> to rbl/rhsbl answers (regexp) if not specified in the ruleset.

        --cacheid <item>, <item>, ...
        This csv-separated list of request attributes will be used to construct
        the request cache identifier. Use this only, if you know exactly what you
        are doing. If you, for example, use postfwd3 only for RBL/RHSBL control,
        you may set this to
                postfwd3 --cache=3600 --cacheid=client_name,client_address
        This increases efficiency of caching and improves postfwd's performance.
        Warning: You should list all items here, which are used in your ruleset!

        --cacheid_md5, --nocacheid_md5    (default=1)
        The cacheid will be created with a md5sum of the request items.

        --cleanup-requests <interval>    (default=600)
        The request cache will be searched for timed out items after this <interval> in
        seconds. It is a minimum value. The cleanup process will only take place, when
        a new request arrives.

        --cleanup-rbls <interval>    (default=600)
        The rbl cache will be searched for timed out items after this <interval> in
        seconds. It is a minimum value. The cleanup process will only take place, when
        a new request arrives.

        --cleanup-rates <interval>    (default=600)
        The rate cache will be searched for timed out items after this <interval> in
        seconds. It is a minimum value. The cleanup process will only take place, when
        a new request arrives.

        -S, --summary <int>    (default=600)
        Shows some usage statistics (program uptime, request counter, matching rules)
        every <int> seconds. This option is included by the -v switch.
        This feature uses the alarm signal, so you can force postfwd3 to dump the stats
        using `kill -ALRM <pid>` (where <pid> is the process id of postfwd).

        Example:
        Aug 19 12:39:45 mail1 postfwd[666]: [STATS] Counters: 213000 seconds uptime, 39 rules
        Aug 19 12:39:45 mail1 postfwd[666]: [STATS] Requests: 71643 overall, 49 last interval, 62.88% cache hits
        Aug 19 12:39:45 mail1 postfwd[666]: [STATS] Averages: 20.18 overall, 4.90 last interval, 557.30 top
        Aug 19 12:39:45 mail1 postfwd[666]: [STATS] Contents: 44 cached requests, 239 cached dnsbl results
        Aug 19 12:39:45 mail1 postfwd[666]: [STATS] Rule ID: R-001   matched: 2704 times
        Aug 19 12:39:45 mail1 postfwd[666]: [STATS] Rule ID: R-002   matched: 9351 times
        Aug 19 12:39:45 mail1 postfwd[666]: [STATS] Rule ID: R-003   matched: 3116 times
        ...

        --no-rulestats
        Disables per rule statistics. Keeps your log clean, if you do not use them.
        This option has no effect without --summary or --verbose set.

        -L, --stdout
        Redirects all syslog messages to stdout for debugging. Do not use this in daemon mode!

       --stdin
        Tells postfwd to get the request data from STDIN instead of a network socket. This
        may be used to test rulesets at the command-line:

        Example:
        postfwd -f /etc/postfwd/postfwd.cf --stdin --stdout --nodaemon ../tools/request.sample

        --cmd
        Shorthand for the combination --stdin, --stdout and --nodaemon

        Example:
        postfwd -f /etc/postfwd/postfwd.cf --cmd ../tools/request.sample

        -t, --test
        In test mode postfwd3 always returns "dunno", but logs according
        to it`s ruleset. -v will be set automatically with this option.

        -n, --nodns
        Disables all DNS based checks like RBL checks. Rules containing
        such elements will be ignored.

        -n, --nodnslog
        Disables logging of dns events.

        --dns_timeout     (default: 14)
        Sets the timeout for asynchonous dns queries in seconds. This value will apply to
        all dns items in a rule.

        --dns_timeout_max    (default: 10)
        Sets the maximum timeout counter for dnsbl lookups. If the timeouts exceed this value
        the corresponding dnsbl will be deactivated for a while (see --dns_timeout_interval).

        --dns_timeout_interval    (default=1200)
        The dnsbl timeout counter will be cleaned after this interval in seconds. Use this
        in conjunction with the --dns_timeout_max parameter.

        --dns_async_txt
        Perform dnsbl A and TXT lookups simultaneously (otherwise only for listings with at
        least one A record). This needs more network bandwidth due to increased queries but
        might increase throughput because the lookups can be parallelized.

        --dns_max_ns_lookups     (default=0)
        maximum ns names to lookup up with sender_ns_addrs item. use 0 for no maximum.

        --dns_max_mx_lookups     (default=0)
        maximum mx names to lookup up with sender_mx_addrs item. use 0 for no maximum.

        --ipv6_dnsbl            (default=0)
        enables dnsbl checks for IPv6 addresses

        -I, --instantcfg
        The config files, specified by -f will be re-read for every request
        postfwd3 receives. This enables on-the-fly configuration changes
        without restarting. Though files will be read only if necessary
        (which means their access times changed since last read) this might
        significantly increase system load.

        --config_timeout    (default=3)
        timeout in seconds to parse a single configuration line. if exceeded, the rule will
        be skipped. this is used to prevent problems due to large files or loops.
        
        --keep_groups    (default=0)
        With this option set postfwd3 does not clear the group cache on reload. Please
        note that you have to restart (not reload) postfwd with this option if you change
        any group based rules.

        --save_groups    (default=none)
        With this option postfwd saves existing groups to disk and reloads them on program
        start. This allows persistent rate limits across program restarts or reboots.
        Please note that postfwd needs read and write access to the specified file.

        --keep_rates    (default=0)
        With this option set postfwd3 does not clear the rate limit counters on reload. Please
        note that you have to restart (not reload) postfwd with this option if you change
        any rate limit rules.

        --save_rates    (default=none)
        With this option postfwd saves existing rate limit counters to disk and reloads them
        on program start. This allows persistent rate limits across program restarts or reboots.
        Please note that postfwd needs read and write access to the specified file.

        -A, --aggregate_addrs, --noaggregate_addrs
        Pre-computes ip address lists to subnets, so that e.g.:

           client_address=10.0.0.0/24, 10.0.1.0/24, 10.0.2.0/24, 10.0.3.0/24
                will result in
           client_address=10.0.0.0/22

        This increases the performance of cidr-matching

        --no_netaddr
        Do not use functions from module NetAddr::IP, even if it was found
        on the system. This option implicitly disables --aggregate_addrs.

        --no_netcidr
        Do not use functions from module Net::CIDR::Lite, even if it was found
        on the system.

        --cidr_method=s   (default=autodetect)
        Use method <s> for network checks. Valid arguments are:
          netcidr       - compare networks using Net::CIDR::Lite Module (v4 and v6 cidr)
          netaddr       - compare networks using NetAddr::IP Module (v4 and v6 cidr)
          postfwd       - old but very fast method (v4=cidr, v6=regex)
        If not specified postfwd tries to load modules in the above order and falls back
        to 'postfwd' if nothing found.

Informational arguments

These arguments are for command line usage only. Never ever use them with postfix!

        -C, --showconfig
        Displays the current ruleset. Use -v for verbose output.

        -V, --version
        Displays the program version.

        -h, --help
        Shows program usage.

        -m, --manual
        Displays the program manual.

        -D, --defaults
        displays complete postfwd3 settings.

        -P, --perfmon
        This option turns of any syslogging and output. It is included
        for performance testing.

        --dumpstats
        Displays program usage statistics.

        --dumpcache
        Displays cache contents.

        --delcache <item>
        Removes an item from the request cache. Use --dumpcache to identify objects.
        E.g.:
                # postfwd --dumpcache
                ...
                %rate_cache -> %sender=gmato@jqvo.org -> %RATE002+2_600 -> @count    -> '1'
                %rate_cache -> %sender=gmato@jqvo.org -> %RATE002+2_600 -> @maxcount -> '2'
                ...
                # postfwd --delrate="sender=gmato@jqvo.org"
                rate cache item 'sender=gmato@jqvo.org' removed

        --delrate <item>
        Removes an item from the rate cache. Use --dumpcache to identify objects.

REFRESH

In daemon mode postfwd3 reloads it's ruleset after receiving a HUP signal. Please see the description of the '-I' switch to have your configuration refreshed for every request postfwd3 receives.

EXAMPLES

        ## whitelisting
        # 1. networks 192.168.1.0/24, 192.168.2.4
        # 2. client_names *.gmx.net and *.gmx.de
        # 3. sender *@someshop.tld from 11.22.33.44
        id=WL001; action=dunno ; client_address=192.168.1.0/24, 192.168.2.4
        id=WL002; action=dunno ; client_name=\.gmx\.(net|de)$
        id=WL003; action=dunno ; sender=@someshop\.tld$ ; client_address=11.22.33.44

        ## TLS control
        # 1. *@authority.tld only with correct TLS fingerprint
        # 2. *@secret.tld only with keysizes >=64
        id=TL001; action=dunno                          ; sender=@authority\.tld$ ; ccert_fingerprint=AA:BB:CC..
        id=TL002; action=REJECT wrong TLS fingerprint   ; sender=@authority\.tld$
        id=TL003; action=REJECT tls keylength < 64      ; sender=@secret\.tld$ ; encryption_keysize=64

        ## Combined RBL checks
        # This will reject mail if
        # 1. listed on ix.dnsbl.manitu.net
        # 2. listed on zen.spamhaus.org (sbl and xbl, dns cache timeout 1200s instead of 3600s)
        # 3. listed on min 2 of bl.spamcop.net, list.dsbl.org, dnsbl.sorbs.net
        # 4. listed on bl.spamcop.net and one of rhsbl.ahbl.org, rhsbl.sorbs.net
        id=RBL01 ; action=REJECT listed on ix.dnsbl.manitu.net  ; rbl=ix.dnsbl.manitu.net
        id=RBL02 ; action=REJECT listed on zen.spamhaus.org     ; rbl=zen.spamhaus.org/127.0.0.[2-8]/1200
        id=RBL03 ; action=REJECT listed on too many RBLs        ; rblcount=2 ; rbl=bl.spamcop.net, list.dsbl.org, dnsbl.sorbs.net
        id=RBL04 ; action=REJECT combined RBL+RHSBL check       ; rbl=bl.spamcop.net ; rhsbl=rhsbl.ahbl.org, rhsbl.sorbs.net

        ## Message size (requires message_size_limit to be set to 30000000)
        # 1. 30MB for systems in *.customer1.tld
        # 2. 20MB for SASL user joejob
        # 3. 10MB default
        id=SZ001; protocol_state==END-OF-MESSAGE; action=DUNNO; size<=30000000 ; client_name=\.customer1.tld$
        id=SZ002; protocol_state==END-OF-MESSAGE; action=DUNNO; size<=20000000 ; sasl_username==joejob
        id=SZ002; protocol_state==END-OF-MESSAGE; action=DUNNO; size<=10000000
        id=SZ100; protocol_state==END-OF-MESSAGE; action=REJECT message too large

        ## Selective Greylisting
        ##
        ## Note that postfwd does not include greylisting. This setup requires a running postgrey service
        ## at port 10031 and the following postfix restriction class in your main.cf:
        ##
        ##      smtpd_restriction_classes = check_postgrey, ...
        ##      check_postgrey = check_policy_service inet:127.0.0.1:10031
        #
        # 1. if listed on zen.spamhaus.org with results 127.0.0.10 or .11, dns cache timeout 1200s
        # 2. Client has no rDNS
        # 3. Client comes from several dialin domains
        id=GR001; action=check_postgrey ; rbl=dul.dnsbl.sorbs.net, zen.spamhaus.org/127.0.0.1[01]/1200
        id=GR002; action=check_postgrey ; client_name=^unknown$
        id=GR003; action=check_postgrey ; client_name=\.(t-ipconnect|alicedsl|ish)\.de$

        ## Date Time
        date=24.12.2007-26.12.2007          ;  action=450 4.7.1 office closed during christmas
        time=04:00:00-05:00:00              ;  action=450 4.7.1 maintenance ongoing, try again later
        time=-07:00:00 ;  sasl_username=jim ;  action=450 4.7.1 to early for you, jim
        time=22:00:00- ;  sasl_username=jim ;  action=450 4.7.1 to late now, jim
        months=-Apr                         ;  action=450 4.7.1 see you in may
        days=!!Mon-Fri                      ;  action=check_postgrey

        ## Usage of jump
        # The following allows a message size of 30MB for different
        # users/clients while others will only have 10MB.
        id=R001 ; action=jump(R100) ; sasl_username=^(Alice|Bob|Jane)$
        id=R002 ; action=jump(R100) ; client_address=192.168.1.0/24
        id=R003 ; action=jump(R100) ; ccert_fingerprint=AA:BB:CC:DD:...
        id=R004 ; action=jump(R100) ; ccert_fingerprint=AF:BE:CD:DC:...
        id=R005 ; action=jump(R100) ; ccert_fingerprint=DD:CC:BB:DD:...
        id=R099 ; protocol_state==END-OF-MESSAGE; action=REJECT message too big (max. 10MB); size=10000000
        id=R100 ; protocol_state==END-OF-MESSAGE; action=REJECT message too big (max. 30MB); size=30000000

        ## Usage of score
        # The following rejects a mail, if the client
        # - is listed on 1 RBL and 1 RHSBL
        # - is listed in 1 RBL or 1 RHSBL and has no correct rDNS
        # - other clients without correct rDNS will be greylist-checked
        # - some whitelists are used to lower the score
        id=S01 ; score=2.6              ; action=check_postgrey
        id=S02 ; score=5.0              ; action=REJECT postfwd score too high
        id=R00 ; action=score(-1.0)     ; rbl=exemptions.ahbl.org,list.dnswl.org,query.bondedsender.org,spf.trusted-forwarder.org
        id=R01 ; action=score(2.5)      ; rbl=bl.spamcop.net, list.dsbl.org, dnsbl.sorbs.net
        id=R02 ; action=score(2.5)      ; rhsbl=rhsbl.ahbl.org, rhsbl.sorbs.net
        id=N01 ; action=score(-0.2)     ; client_name==$$helo_name
        id=N02 ; action=score(2.7)      ; client_name=^unknown$
        ...

        ## Usage of rate and size
        # The following temporary rejects requests from "unknown" clients, if they
        # 1. exceeded 30 requests per hour or
        # 2. tried to send more than 1.5mb within 10 minutes
        id=RATE01 ;  client_name==unknown ;  protocol_state==RCPT
                action=rate(client_address/30/3600/450 4.7.1 sorry, max 30 requests per hour)
        id=SIZE01 ;  client_name==unknown ;  protocol_state==END-OF-MESSAGE
                action=size(client_address/1572864/600/450 4.7.1 sorry, max 1.5mb per 10 minutes)

        ## Macros
        # definition
        &&RBLS { rbl=zen.spamhaus.org,list.dsbl.org,bl.spamcop.net,dnsbl.sorbs.net,ix.dnsbl.manitu.net; };
        &&GONOW { action=REJECT your request caused our spam detection policy to reject this message. More info at http://www.domain.local; };
        # rules
        &&GONOW ;  &&RBLS ;  client_name=^unknown$
        &&GONOW ;  &&RBLS ;  client_name=(\d+[\.-_]){4}
        &&GONOW ;  &&RBLS ;  client_name=[\.-_](adsl|dynamic|ppp|)[\.-_]

        ## Groups
        # definition
        &&RBLS{
                rbl=zen.spamhaus.org
                rbl=list.dsbl.org
                rbl=bl.spamcop.net
                rbl=dnsbl.sorbs.net
                rbl=ix.dnsbl.manitu.net
        };
        &&RHSBLS{
                ...
        };
        &&DYNAMIC{
                client_name==unknown
                client_name~=(\d+[\.-_]){4}
                client_name~=[\.-_](adsl|dynamic|ppp|)[\.-_]
                ...
        };
        &&BAD_HELO{
                helo_name==my.name.tld
                helo_name~=^([^\.]+)$
                helo_name~=\.(local|lan)$
                ...
        };
        &&MAINTENANCE{
                date=15.01.2007
                date=15.04.2007
                date=15.07.2007
                date=15.10.2007
                time=03:00:00 - 04:00:00
        };
        # rules
        id=COMBINED    ;  &&RBLS ;  &&DYNAMIC ;  action=REJECT dynamic client and listed on RBL
        id=MAINTENANCE ;  &&MAINTENANCE       ;  action=DEFER maintenance time - please try again later
        
        # now with the set() command, note that long item
        # lists don't have to be compared twice
        id=RBL01    ;  &&RBLS      ;  action=set(HIT_rbls=1)
        id=HELO01   ;  &&BAD_HELO  ;  action=set(HIT_helo=1)
        id=DYNA01   ;  &&DYNAMIC   ;  action=set(HIT_dyna=1)
        id=REJECT01 ;  HIT_rbls==1 ;  HIT_helo==1  ; action=REJECT please see http://some.org/info?reject=01 for more info
        id=REJECT02 ;  HIT_rbls==1 ;  HIT_dyna==1  ; action=REJECT please see http://some.org/info?reject=02 for more info
        id=REJECT03 ;  HIT_helo==1 ;  HIT_dyna==1  ; action=REJECT please see http://some.org/info?reject=03 for more info

        ## combined with enhanced rbl features
        #
        id=RBL01 ; rhsblcount=all ; rblcount=all ; &&RBLS ; &&RHSBLS
             action=set(HIT_dnsbls=$$rhsblcount,HIT_dnsbls+=$$rblcount,HIT_dnstxt=$$dnsbltext)
        id=RBL02 ; HIT_dnsbls>=2  ; action=554 5.7.1 blocked using $$HIT_dnsbls DNSBLs [INFO: $$HIT_dnstxt]

PARSER

Configuration

The postfwd3 ruleset can be specified at the commandline (-r option) or be read from files (-f). The order of your arguments will be kept. You should check the parser with the -C | --showconfig switch at the command line before applying a new config. The following call:

        postfwd3 --showconfig \
                -r "id=TEST; recipient_count=100; action=WARN mail with 100+ recipients" \
                -f /etc/postfwd.cf \
                -r "id=DEFAULT; action=dunno";

will produce the following output:

        Rule   0: id->"TEST" action->"WARN mail with 100+ recipients"; recipient_count->"100"
        ...
        ... <content of /etc/postfwd.cf> ...
        ...
        Rule <n>: id->"DEFAULT" action->"dunno"

Multiple items of the same type will be added to lists (see the "ITEMS" section for more info):

        postfwd3 --showconfig \
                -r "client_address=192.168.1.0/24; client_address=172.16.26.32; action=dunno"

will result in:

        Rule   0: id->"R-0"; action->"dunno"; client_address->"192.168.1.0/24, 172.16.26.32"

Macros are evaluated at configuration stage, which means that

        postfwd3 --showconfig \
                -r "&&RBLS { rbl=bl.spamcop.net; client_name=^unknown$; };" \
                -r "id=RBL001; &&RBLS; action=REJECT listed on spamcop and bad rdns";

will result in:

        Rule   0: id->"RBL001"; action->"REJECT listed on spamcop and bad rdns"; rbl->"bl.spamcop.net"; client_name->"^unknown$"

Request processing

When a policy delegation request arrives it will be compared against postfwd`s ruleset. To inspect the processing in detail you should increase verbority using use the "-v" or "-vv" switch. "-L" redirects log messages to stdout.

Keeping the order of the ruleset in general, items will be compared in random order, which basically means that

        id=R001; action=dunno; client_address=192.168.1.1; sender=bob@alice.local

equals to

        id=R001; sender=bob@alice.local; client_address=192.168.1.1; action=dunno

Lists will be evaluated in the specified order. This allows to place faster expressions at first:

        postfwd3 -vv --cmd -r "id=RBL001; rbl=localrbl.local zen.spamhaus.org; action=REJECT" /some/where/request.sample

produces the following

        [LOGS info]: compare rbl: "remotehost.remote.net[68.10.1.7]"  ->  "localrbl.local"
        [LOGS info]: count1 rbl:  "2"  ->  "0"
        [LOGS info]: query rbl:   localrbl.local 7.1.10.68 (7.1.10.68.localrbl.local)
        [LOGS info]: count2 rbl:  "2"  ->  "0"
        [LOGS info]: match rbl:   FALSE
        [LOGS info]: compare rbl: "remotehost.remote.net[68.10.1.7]"  ->  "zen.spamhaus.org"
        [LOGS info]: count1 rbl:  "2"  ->  "0"
        [LOGS info]: query rbl:   zen.spamhaus.org 7.1.10.68 (7.1.10.68.zen.spamhaus.org)
        [LOGS info]: count2 rbl:  "2"  ->  "0"
        [LOGS info]: match rbl:   FALSE
        [LOGS info]: Action: dunno

The negation operator !!(<value>) has the highest priority and therefore will be evaluated first. Then variable substitutions are performed:

        postfwd3 -vv --cmd -r "id=TEST; action=REJECT; client_name=!!($$heloname)" /some/where/request.sample

will give

        [LOGS info]: compare client_name:     "unknown"  ->  "!!($$helo_name)"
        [LOGS info]: negate client_name:      "unknown"  ->  "$$helo_name"
        [LOGS info]: substitute client_name:  "unknown"  ->  "english-breakfast.cloud8.net"
        [LOGS info]: match client_name:  TRUE
        [LOGS info]: Action: REJECT

Ruleset evaluation

A rule hits when all items (or at least one element of a list for each item) have matched. As soon as one item (or all elements of a list) fails to compare against the request attribute the parser will jump to the next rule in the postfwd3 ruleset.

If a rule matches, there are two options:

* Rule returns postfix action (dunno, reject, ...) The parser stops rule processing and returns the action to postfix. Other rules will not be evaluated.

* Rule returns postfwd3 action (jump(), note(), ...) The parser evaluates the given action and continues with the next rule (except for the jump() or quit() actions - please see the "ACTIONS" section for more information). Nothing will be sent to postfix.

If no rule has matched and the end of the ruleset is reached postfwd3 will return dunno without logging anything unless in verbose mode. You may place a last catch-all rule to change that behaviour:

        ... <your rules> ...
        id=DEFAULT ;  action=dunno

will log any request that passes the ruleset without having hit a prior rule.

SETTINGS

Since version postfwd3 2.00-pre6 the program settings can be retrieved from a file. The information must be parseable by perl eval() function. Comments are allowed:

       # example settings for postfwd3
       {
          # enable verbose logging
          verbose => 1,

          # user and group for postfwd
          base => {
              group => 'postfw',
              user => 'postfw'
          },

          # server settings
          server => {
              proto => 'tcp',
              host => '127.0.0.1',
              port => '10099'
          }
       }

postfwd3 can also export the program settings in that format using Data::Dumper.

        # show configuration
        postfwd3 --showsettings

        # save configuration to file
        postfwd3 --savesettings=/path/to/file

        # read configuration from file
        postfwd3 --loadsettings=/path/to/file

Multiple usage of --loadsettings or the short form -F is allowed. The order will be kept. Commandline arguments override the settings retrieved from a file.

        # load base settings, override with node settings and set port and verbose logging
        postfwd3 -v -p 10050 -F /path/to/basefile -F /path/to/nodefile

Samples for settings files are distributed in the etc/-folder of the postfwd3 tarball.

DEBUGGING

To debug special steps of the parser the '--debug' switch takes a list of debug classes. Since postfwd3 2.00 the list of available debug classes can be retrieved by running:

        postfwd3 --debugclasses

INTEGRATION

Integration via daemon mode

The common way to use postfwd3 is to start it as daemon, listening at a specified tcp port. postfwd3 will spawn multiple child processes which communicate with a parent cache. This is the prefered way to use postfwd3 in high volume environments. Start postfwd3 with the following parameters:

        postfwd3 -d -f /etc/postfwd.cf -i 127.0.0.1 -p 10045 -u nobody -g nobody -S

For efficient caching you should check if you can use the options --cacheid, --cache-rdomain-only, --cache-no-sender and --cache-no-size.

Now check your syslogs (default facility "mail") for a line like:

        Aug  9 23:00:24 mail postfwd[5158]: postfwd3 n.nn ready for input

and use `netstat -an|grep 10045` to check for something like

        tcp  0  0  127.0.0.1:10045  0.0.0.0:*  LISTEN

If everything works, open your postfix main.cf and insert the following

        127.0.0.1:10045_time_limit      = 3600                                          <--- integration
        smtpd_recipient_restrictions    = permit_mynetworks                             <--- recommended
                                          reject_unauth_destination                     <--- recommended
                                          check_policy_service inet:127.0.0.1:10045     <--- integration

Reload your configuration with `postfix reload` and watch your logs. In it works you should see lines like the following in your mail log:

        Aug  9 23:01:24 mail postfwd[5158]: rule=22, id=ML_POSTFIX, client=english-breakfast.cloud9.net[168.100.1.7], sender=owner-postfix-users@postfix.tld, recipient=someone@domain.local, helo=english-breakfast.cloud9.net, proto=ESMTP, state=RCPT, action=dunno

If you want to check for size or rcpt_count items you must integrate postfwd3 in smtp_data_restrictions or smtpd_end_of_data_restrictions. Of course you can also specify a restriction class and use it in your access tables. First create a file /etc/postfix/policy containing:

        domain1.local           postfwdcheck
        domain2.local           postfwdcheck
        ...

Then postmap that file (`postmap hash:/etc/postfix/policy`), open your main.cf and enter

        # Restriction Classes
        smtpd_restriction_classes       = postfwdcheck, <some more>...                          <--- integration
        postfwdcheck                    = check_policy_service inet:127.0.0.1:10045             <--- integration

        127.0.0.1:10045_time_limit      = 3600                                                  <--- integration
        smtpd_recipient_restrictions    = permit_mynetworks,                                    <--- recommended
                                          reject_unauth_destination,                            <--- recommended
                                          ...                                                   <--- optional
                                          check_recipient_access hash:/etc/postfix/policy,      <--- integration
                                          ...                                                   <--- optional

Reload postfix and watch your logs.

Integration via docker

postfwd can be run in a docker container. The relevant options are --nodaemon and --stdout. More information can be found in the included doc/docker.html or at http://postfwd.org/docker.

TESTING

First you have to create a ruleset (see Configuration section). Check it with

        postfwd3 -f /etc/postfwd.cf -C

There is an example policy request distributed with postfwd, called 'request.sample'. Simply change it to meet your requirements and use

        postfwd3 --cmd -f /etc/postfwd.cf request.sample

You should get an answer like

        action=<whateveryouconfigured>

For network tests I use netcat:

        nc 127.0.0.1 10045 <request.sample

to send a request to postfwd. If you receive nothing, make sure that postfwd3 is running and listening on the specified network settings.

PERFORMANCE

Some of these proposals might not match your environment. Please check your requirements and test new options carefully!

        - use caching options
        - use the correct match operator ==, <=, >=
        - use ^ and/or $ in regular expressions
        - use item lists (faster than single rules)
        - use set() action on repeated item lists
        - use jumps and rate limits
        - use a pre-lookup rule for rbl/rhsbls with empty note() action

SEE ALSO

See http://www.postfix.org/SMTPD_POLICY_README.html for a description of how Postfix policy servers work.

LICENSE

postfwd3 is free software and released under BSD license, which basically means that you can do what you want as long as you keep the copyright notice:

Copyright (c) 2009, Jan Peter Kessler All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

 * Redistributions of source code must retain the above copyright
   notice, this list of conditions and the following disclaimer.
 * Redistributions in binary form must reproduce the above copyright
   notice, this list of conditions and the following disclaimer in
   the documentation and/or other materials provided with the
   distribution.
 * Neither the name of the authors nor the names of his contributors
   may be used to endorse or promote products derived from this
   software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY ME ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

AUTHOR

Jan Peter Kessler <info (AT) postfwd (DOT) org>. Let me know, if you have any suggestions.