1 ######################################################################
3 # As of 2.0.0, FreeRADIUS supports virtual hosts using the
4 # "server" section, and configuration directives.
6 # Virtual hosts should be put into the "sites-available"
7 # directory. Soft links should be created in the "sites-enabled"
8 # directory to these files. This is done in a normal installation.
10 # If you are using 802.1X (EAP) authentication, please see also
11 # the "inner-tunnel" virtual server. You will likely have to edit
12 # that, too, for authentication to work.
16 ######################################################################
18 # Read "man radiusd" before editing this file. See the section
19 # titled DEBUGGING. It outlines a method where you can quickly
20 # obtain the configuration you want, without running into
21 # trouble. See also "man unlang", which documents the format
24 # This configuration is designed to work in the widest possible
25 # set of circumstances, with the widest possible number of
26 # authentication methods. This means that in general, you should
27 # need to make very few changes to this file.
29 # The best way to configure the server for your local system
30 # is to CAREFULLY edit this file. Most attempts to make large
31 # edits to this file will BREAK THE SERVER. Any edits should
32 # be small, and tested by running the server with "radiusd -X".
33 # Once the edits have been verified to work, save a copy of these
34 # configuration files somewhere. (e.g. as a "tar" file). Then,
35 # make more edits, and test, as above.
37 # There are many "commented out" references to modules such
38 # as ldap, sql, etc. These references serve as place-holders.
39 # If you need the functionality of that module, then configure
40 # it in radiusd.conf, and un-comment the references to it in
41 # this file. In most cases, those small changes will result
42 # in the server being able to connect to the DB, and to
45 ######################################################################
49 # If you want the server to listen on additional addresses, or on
50 # additional ports, you can use multiple "listen" sections.
52 # Each section make the server listen for only one type of packet,
53 # therefore authentication and accounting have to be configured in
56 # The server ignore all "listen" section if you are using '-i' and '-p'
57 # on the command line.
60 # Type of packets to listen for.
62 # auth listen for authentication packets
63 # acct listen for accounting packets
64 # proxy IP to use for sending proxied packets
65 # detail Read from the detail file. For examples, see
66 # raddb/sites-available/copy-acct-to-home-server
67 # status listen for Status-Server packets. For examples,
68 # see raddb/sites-available/status
69 # coa listen for CoA-Request and Disconnect-Request
70 # packets. For examples, see the file
71 # raddb/sites-available/coa
75 # Note: "type = proxy" lets you control the source IP used for
76 # proxying packets, with some limitations:
78 # * A proxy listener CANNOT be used in a virtual server section.
79 # * You should probably set "port = 0".
80 # * Any "clients" configuration will be ignored.
82 # See also proxy.conf, and the "src_ipaddr" configuration entry
83 # in the sample "home_server" section. When you specify the
84 # source IP address for packets sent to a home server, the
85 # proxy listeners are automatically created.
87 # ipaddr/ipv4addr/ipv6addr - IP address on which to listen.
88 # Out of several options the first one will be used.
91 # IPv4 address (e.g. 1.2.3.4, for ipv4addr/ipaddr)
92 # IPv6 address (e.g. 2001:db8::1, for ipv6addr/ipaddr)
93 # hostname (radius.example.com,
94 # A record for ipv4addr,
95 # AAAA record for ipv6addr,
96 # A or AAAA record for ipaddr)
103 # Port on which to listen.
104 # Allowed values are:
105 # integer port number (1812)
106 # 0 means "use /etc/services for the proper port"
109 # Some systems support binding to an interface, in addition
110 # to the IP address. This feature isn't strictly necessary,
111 # but for sites with many IP addresses on one interface,
112 # it's useful to say "listen on all addresses for eth0".
114 # If your system does not support this feature, you will
115 # get an error if you try to use it.
119 # Per-socket lists of clients. This is a very useful feature.
121 # The name here is a reference to a section elsewhere in
122 # radiusd.conf, or clients.conf. Having the name as
123 # a reference allows multiple sockets to use the same
126 # If this configuration is used, then the global list of clients
127 # is IGNORED for this "listen" section. Take care configuring
128 # this feature, to ensure you don't accidentally disable a
131 # See clients.conf for the configuration of "per_socket_clients".
133 # clients = per_socket_clients
136 # Connection limiting for sockets with "proto = tcp".
138 # This section is ignored for other kinds of sockets.
142 # Limit the number of simultaneous TCP connections to the socket
145 # Setting this to 0 means "no limit"
148 # The per-socket "max_requests" option does not exist.
151 # The lifetime, in seconds, of a TCP connection. After
152 # this lifetime, the connection will be closed.
154 # Setting this to 0 means "forever".
158 # The idle timeout, in seconds, of a TCP connection.
159 # If no packets have been received over the connection for
160 # this time, the connection will be closed.
162 # Setting this to 0 means "no timeout".
164 # We STRONGLY RECOMMEND that you set an idle timeout.
171 # This second "listen" section is for listening on the accounting
180 # clients = per_socket_clients
183 # The number of packets received can be rate limited via the
184 # "max_pps" configuration item. When it is set, the server
185 # tracks the total number of packets received in the previous
186 # second. If the count is greater than "max_pps", then the
187 # new packet is silently discarded. This helps the server
188 # deal with overload situations.
190 # The packets/s counter is tracked in a sliding window. This
191 # means that the pps calculation is done for the second
192 # before the current packet was received. NOT for the current
193 # wall-clock second, and NOT for the previous wall-clock second.
195 # Useful values are 0 (no limit), or 100 to 10000.
196 # Values lower than 100 will likely cause the server to ignore
197 # normal traffic. Few systems are capable of handling more than
200 # It is most useful for accounting systems. Set it to 50%
201 # more than the normal accounting load, and you can be sure that
202 # the server will never get overloaded
206 # Only for "proto = tcp". These are ignored for "udp" sockets.
210 # max_connections = 0
214 # IPv6 versions of the above - read their full config to understand options
217 ipv6addr = :: # any. ::1 == localhost
220 # clients = per_socket_clients
233 # clients = per_socket_clients
239 # max_connections = 0
243 # Authorization. First preprocess (hints and huntgroups files),
244 # then realms, and finally look in the "users" file.
246 # Any changes made here should also be made to the "inner-tunnel"
249 # The order of the realm modules will determine the order that
250 # we try to find a matching realm.
252 # Make *sure* that 'preprocess' comes before any realm if you
253 # need to setup hints for the remote radius server
256 # Take a User-Name, and perform some checks on it, for spaces and other
257 # invalid characters. If the User-Name appears invalid, reject the
260 # See policy.d/filter for the definition of the filter_username policy.
265 # The preprocess module takes care of sanitizing some bizarre
266 # attributes in the request, and turning them into attributes
267 # which are more standard.
269 # It takes care of processing the 'raddb/hints' and the
270 # 'raddb/huntgroups' files.
273 # If you intend to use CUI and you require that the Operator-Name
274 # be set for CUI generation and you want to generate CUI also
275 # for your local clients then uncomment the operator-name
276 # below and set the operator-name for your clients in clients.conf
280 # If you want to generate CUI for some clients that do not
281 # send proper CUI requests, then uncomment the
282 # cui below and set "add_cui = yes" for these clients in clients.conf
286 # If you want to have a log of authentication requests,
287 # un-comment the following line.
291 # The chap module will set 'Auth-Type := CHAP' if we are
292 # handling a CHAP request and Auth-Type has not already been set
296 # If the users are logging in with an MS-CHAP-Challenge
297 # attribute for authentication, the mschap module will find
298 # the MS-CHAP-Challenge attribute, and add 'Auth-Type := MS-CHAP'
299 # to the request, which will cause the server to then use
300 # the mschap module for authentication.
304 # If you have a Cisco SIP server authenticating against
305 # FreeRADIUS, uncomment the following line, and the 'digest'
306 # line in the 'authenticate' section.
310 # The WiMAX specification says that the Calling-Station-Id
311 # is 6 octets of the MAC. This definition conflicts with
312 # RFC 3580, and all common RADIUS practices. Un-commenting
313 # the "wimax" module here means that it will fix the
314 # Calling-Station-Id attribute to the normal format as
315 # specified in RFC 3580 Section 3.21
319 # Look for IPASS style 'realm/', and if not found, look for
320 # '@realm', and decide whether or not to proxy, based on
325 # If you are using multiple kinds of realms, you probably
326 # want to set "ignore_null = yes" for all of them.
327 # Otherwise, when the first style of realm doesn't match,
328 # the other styles won't be checked.
334 # This module takes care of EAP-MD5, EAP-TLS, and EAP-LEAP
337 # It also sets the EAP-Type attribute in the request
338 # attribute list to the EAP type from the packet.
340 # As of 2.0, the EAP module returns "ok" in the authorize stage
341 # for TTLS and PEAP. In 1.x, it never returned "ok" here, so
342 # this change is compatible with older configurations.
344 # The example below uses module failover to avoid querying all
345 # of the following modules if the EAP module returns "ok".
346 # Therefore, your LDAP and/or SQL servers will not be queried
347 # for the many packets that go back and forth to set up TTLS
348 # or PEAP. The load on those servers will therefore be reduced.
355 # Pull crypt'd passwords from /etc/passwd or /etc/shadow,
356 # using the system API's to get the password. If you want
357 # to read /etc/passwd or /etc/shadow directly, see the
358 # passwd module in radiusd.conf.
363 # Read the 'users' file
367 # Look in an SQL database. The schema of the database
368 # is meant to mirror the "users" file.
370 # See "Authorization Queries" in sql.conf
374 # If you are using /etc/smbpasswd, and are also doing
375 # mschap authentication, the un-comment this line, and
376 # configure the 'smbpasswd' module.
380 # The ldap module reads passwords from the LDAP database.
384 # Enforce daily limits on time spent logged in.
392 # If no other module has claimed responsibility for
393 # authentication, then try to use PAP. This allows the
394 # other modules listed above to add a "known good" password
395 # to the request, and to do nothing else. The PAP module
396 # will then see that password, and use it to do PAP
399 # This module should be listed last, so that the other modules
400 # get a chance to set Auth-Type for themselves.
405 # If "status_server = yes", then Status-Server messages are passed
406 # through the following section, and ONLY the following section.
407 # This permits you to do DB queries, for example. If the modules
408 # listed here return "fail", then NO response is sent.
410 # Autz-Type Status-Server {
419 # This section lists which modules are available for authentication.
420 # Note that it does NOT mean 'try each module in order'. It means
421 # that a module from the 'authorize' section adds a configuration
422 # attribute 'Auth-Type := FOO'. That authentication type is then
423 # used to pick the appropriate module from the list below.
426 # In general, you SHOULD NOT set the Auth-Type attribute. The server
427 # will figure it out on its own, and will do the right thing. The
428 # most common side effect of erroneously setting the Auth-Type
429 # attribute is that one authentication method will work, but the
432 # The common reasons to set the Auth-Type attribute by hand
433 # is to either forcibly reject the user (Auth-Type := Reject),
434 # or to or forcibly accept the user (Auth-Type := Accept).
436 # Note that Auth-Type := Accept will NOT work with EAP.
438 # Please do not put "unlang" configurations into the "authenticate"
439 # section. Put them in the "post-auth" section instead. That's what
440 # the post-auth section is for.
444 # PAP authentication, when a back-end database listed
445 # in the 'authorize' section supplies a password. The
446 # password can be clear-text, or encrypted.
452 # Most people want CHAP authentication
453 # A back-end database listed in the 'authorize' section
454 # MUST supply a CLEAR TEXT password. Encrypted passwords
461 # MSCHAP authentication.
467 # If you have a Cisco SIP server authenticating against
468 # FreeRADIUS, uncomment the following line, and the 'digest'
469 # line in the 'authorize' section.
473 # Pluggable Authentication Modules.
476 # Uncomment it if you want to use ldap for authentication
478 # Note that this means "check plain-text password against
479 # the ldap database", which means that EAP won't work,
480 # as it does not supply a plain-text password.
482 # We do NOT recommend using this. LDAP servers are databases.
483 # They are NOT authentication servers. FreeRADIUS is an
484 # authentication server, and knows what to do with authentication.
485 # LDAP servers do not.
492 # Allow EAP authentication.
496 # The older configurations sent a number of attributes in
497 # Access-Challenge packets, which wasn't strictly correct.
498 # If you want to filter out these attributes, uncomment
499 # the following lines.
505 # if (handled && (Response-Packet-Type == Access-Challenge)) {
506 # attr_filter.access_challenge.post-auth
507 # handled # override the "updated" code from attr_filter
514 # Pre-accounting. Decide which accounting type to use.
520 # Merge Acct-[Input|Output]-Gigawords and Acct-[Input-Output]-Octets
521 # into a single 64bit counter Acct-[Input|Output]-Octets64.
526 # Session start times are *implied* in RADIUS.
527 # The NAS never sends a "start time". Instead, it sends
528 # a start packet, *possibly* with an Acct-Delay-Time.
529 # The server is supposed to conclude that the start time
530 # was "Acct-Delay-Time" seconds in the past.
532 # The code below creates an explicit start time, which can
533 # then be used in other modules. It will be *mostly* correct.
534 # Any errors are due to the 1-second resolution of RADIUS,
535 # and the possibility that the time on the NAS may be off.
537 # The start time is: NOW - delay - session_length
541 # FreeRADIUS-Acct-Session-Start-Time = "%{expr: %l - %{%{Acct-Session-Time}:-0} - %{%{Acct-Delay-Time}:-0}}"
546 # Ensure that we have a semi-unique identifier for every
547 # request, and many NAS boxes are broken.
551 # Look for IPASS-style 'realm/', and if not found, look for
552 # '@realm', and decide whether or not to proxy, based on
555 # Accounting requests are generally proxied to the same
556 # home server as authentication requests.
562 # Read the 'acct_users' file
567 # Accounting. Log the accounting data.
570 # Update accounting packet by adding the CUI attribute
571 # recorded from the corresponding Access-Accept
572 # use it only if your NAS boxes do not support CUI themselves
575 # Create a 'detail'ed log of the packets.
576 # Note that accounting requests which are proxied
577 # are also logged in the detail file.
581 # Update the wtmp file
583 # If you don't use "radlast", you can delete this line.
587 # For Simultaneous-Use tracking.
589 # Due to packet losses in the network, the data here
590 # may be incorrect. There is little we can do about it.
594 # Return an address to the IP Pool when we see a stop record.
598 # Log traffic to an SQL database.
600 # See "Accounting queries" in sql.conf
604 # If you receive stop packets with zero session length,
605 # they will NOT be logged in the database. The SQL module
606 # will print a message (only in debugging mode), and will
609 # You can ignore these packets by uncommenting the following
610 # three lines. Otherwise, the server will not respond to the
611 # accounting request, and the NAS will retransmit.
618 # Instead of sending the query to the SQL server,
619 # write it into a log file.
623 # Cisco VoIP specific bulk accounting
626 # For Exec-Program and Exec-Program-Wait
629 # Filter attributes from the accounting response.
630 attr_filter.accounting_response
633 # See "Autz-Type Status-Server" for how this works.
635 # Acct-Type Status-Server {
641 # Session database, used for checking Simultaneous-Use. Either the radutmp
642 # or rlm_sql module can handle this.
643 # The rlm_sql module is *much* faster
648 # See "Simultaneous Use Checking Queries" in sql.conf
653 # Post-Authentication
654 # Once we KNOW that the user has been authenticated, there are
655 # additional steps we can take.
658 # For EAP-TTLS and PEAP, add the cached attributes to the reply.
659 # The "session-state" attributes are automatically cached when
660 # an Access-Challenge is sent, and automatically retrieved
661 # when an Access-Request is received.
663 # The session-state attributes are automatically deleted after
664 # an Access-Reject or Access-Accept is sent.
667 &reply: += &session-state:
670 # Get an address from the IP Pool.
674 # Create the CUI value and add the attribute to Access-Accept.
675 # Uncomment the line below if *returning* the CUI.
679 # If you want to have a log of authentication replies,
680 # un-comment the following line, and enable the
681 # 'detail reply_log' module.
685 # After authenticating the user, do another SQL query.
687 # See "Authentication Logging Queries" in sql.conf
691 # Instead of sending the query to the SQL server,
692 # write it into a log file.
697 # Un-comment the following if you want to modify the user's object
698 # in LDAP after a successful login.
702 # For Exec-Program and Exec-Program-Wait
706 # Calculate the various WiMAX keys. In order for this to work,
707 # you will need to define the WiMAX NAI, usually via
710 # WiMAX-MN-NAI = "%{User-Name}"
713 # If you want various keys to be calculated, you will need to
714 # update the reply with "template" values. The module will see
715 # this, and replace the template values with the correct ones
716 # taken from the cryptographic calculations. e.g.
719 # WiMAX-FA-RK-Key = 0x00
720 # WiMAX-MSK = "%{EAP-MSK}"
723 # You may want to delete the MS-MPPE-*-Keys from the reply,
724 # as some WiMAX clients behave badly when those attributes
725 # are included. See "raddb/modules/wimax", configuration
726 # entry "delete_mppe_keys" for more information.
731 # If there is a client certificate (EAP-TLS, sometimes PEAP
732 # and TTLS), then some attributes are filled out after the
733 # certificate verification has been performed. These fields
734 # MAY be available during the authentication, or they may be
735 # available only in the "post-auth" section.
737 # The first set of attributes contains information about the
738 # issuing certificate which is being used. The second
739 # contains information about the client certificate (if
743 # Reply-Message += "%{TLS-Cert-Serial}"
744 # Reply-Message += "%{TLS-Cert-Expiration}"
745 # Reply-Message += "%{TLS-Cert-Subject}"
746 # Reply-Message += "%{TLS-Cert-Issuer}"
747 # Reply-Message += "%{TLS-Cert-Common-Name}"
748 # Reply-Message += "%{TLS-Cert-Subject-Alt-Name-Email}"
750 # Reply-Message += "%{TLS-Client-Cert-Serial}"
751 # Reply-Message += "%{TLS-Client-Cert-Expiration}"
752 # Reply-Message += "%{TLS-Client-Cert-Subject}"
753 # Reply-Message += "%{TLS-Client-Cert-Issuer}"
754 # Reply-Message += "%{TLS-Client-Cert-Common-Name}"
755 # Reply-Message += "%{TLS-Client-Cert-Subject-Alt-Name-Email}"
758 # Insert class attribute (with unique value) into response,
759 # aids matching auth and acct records, and protects against duplicate
760 # Acct-Session-Id. Note: Only works if the NAS has implemented
761 # RFC 2865 behaviour for the class attribute, AND if the NAS
762 # supports long Class attributes. Many older or cheap NASes
763 # only support 16-octet Class attributes.
766 # MacSEC requires the use of EAP-Key-Name. However, we don't
767 # want to send it for all EAP sessions. Therefore, the EAP
768 # modules put required data into the EAP-Session-Id attribute.
769 # This attribute is never put into a request or reply packet.
771 # Uncomment the next few lines to copy the required data into
772 # the EAP-Key-Name attribute
773 # if (reply:EAP-Session-Id) {
775 # EAP-Key-Name := "%{reply:EAP-Session-Id}"
779 # Remove reply message if the response contains an EAP-Message
780 remove_reply_message_if_eap
783 # Access-Reject packets are sent through the REJECT sub-section of the
786 # Add the ldap module name (or instance) if you have set
787 # 'edir_account_policy_check = yes' in the ldap module configuration
789 # The "session-state" attributes are not available here.
791 Post-Auth-Type REJECT {
792 # log failed authentications in SQL, too.
794 attr_filter.access_reject
796 # Insert EAP-Failure message if the request was
797 # rejected by policy instead of because of an
798 # authentication failure
801 # Remove reply message if the response contains an EAP-Message
802 remove_reply_message_if_eap
807 # When the server decides to proxy a request to a home server,
808 # the proxied request is first passed through the pre-proxy
809 # stage. This stage can re-write the request, or decide to
812 # Only a few modules currently have this method.
815 # Before proxing the request add an Operator-Name attribute identifying
816 # if the operator-name is found for this client.
817 # No need to uncomment this if you have already enabled this in
818 # the authorize section.
821 # The client requests the CUI by sending a CUI attribute
822 # containing one zero byte.
823 # Uncomment the line below if *requesting* the CUI.
826 # Uncomment the following line if you want to change attributes
827 # as defined in the preproxy_users file.
830 # Uncomment the following line if you want to filter requests
831 # sent to remote servers based on the rules defined in the
832 # 'attrs.pre-proxy' file.
833 # attr_filter.pre-proxy
835 # If you want to have a log of packets proxied to a home
836 # server, un-comment the following line, and the
837 # 'detail pre_proxy_log' section, above.
842 # When the server receives a reply to a request it proxied
843 # to a home server, the request may be massaged here, in the
848 # If you want to have a log of replies from a home server,
849 # un-comment the following line, and the 'detail post_proxy_log'
853 # Uncomment the following line if you want to filter replies from
854 # remote proxies based on the rules defined in the 'attrs' file.
855 # attr_filter.post-proxy
858 # If you are proxying LEAP, you MUST configure the EAP
859 # module, and you MUST list it here, in the post-proxy
862 # You MUST also use the 'nostrip' option in the 'realm'
863 # configuration. Otherwise, the User-Name attribute
864 # in the proxied request will not match the user name
865 # hidden inside of the EAP packet, and the end server will
866 # reject the EAP request.
871 # If the server tries to proxy a request and fails, then the
872 # request is processed through the modules in this section.
874 # The main use of this section is to permit robust proxying
875 # of accounting packets. The server can be configured to
876 # proxy accounting packets as part of normal processing.
877 # Then, if the home server goes down, accounting packets can
878 # be logged to a local "detail" file, for processing with
879 # radrelay. When the home server comes back up, radrelay
880 # will read the detail file, and send the packets to the
883 # With this configuration, the server always responds to
884 # Accounting-Requests from the NAS, but only writes
885 # accounting packets to disk if the home server is down.
887 # Post-Proxy-Type Fail {