- Freeradius-Users - lists.freeradius.org

How to best configure Radius for my domain ?
by Analyser Great 12 Jul '13

12 Jul '13

I am not sure what I am doing wrong but I am just banging my head on a simple problem. I am trying to upgrade from radius 1 to radius 2 and will be using an entire set of new configuration. Currently, I am checking an Ldap attribute "dialupaccess" to allow users to get access to vpn / network devices. Unfortunately everyone has access to everything in this scenario (Don't ask, I didn't set this up !!) I am trying to setup a configuration where network admins have access to all devices and users have only access to cisco vpn. I don't wanna use local user database since I already have Ldap to authorize and Kerberos to authenticate. How do you do this in your domain currently with version 2 ? Much appreciated John.

2 1

mod_auth_radius w/Apache 2.4.4 ??
by laurence.schuler 12 Jul '13

12 Jul '13

I'm trying to use mod_auth_radius(-2.0) with apache 2.4.4 and it does not appear to be working properly. It complains: [:warn] [pid 14690] AuthRadiusActive set, but no RADIUS server IP - missing AddRadiusAuth in this context?) When I have AuthRadiusAuth set, and I can confirm it by changing the hostname to garbage, the server will then fail to start. So, it seems the module needs to be updated for apache 2.4.4. Is this activity planned? anyone have patches? Thanks, --Larry

2 1

maintaining reSIProcate compatibility with FreeRADIUS
by Daniel Pocock 11 Jul '13

11 Jul '13

Hi, A few years ago, I adapted the RADIUS client code from SER to work in reSIProcate and specifically the SIP proxy, repro I'm now reviewing the code to work out how to extend it for reTurn, the TURN server and to see if any other changes are necessary. Things have changed slightly since reSIProcate originally adopted RADIUS support. The original implementation is based on http://tools.ietf.org/html/draft-sterman-aaa-sip-04 and that works with FreeRADIUS (or it did work at the time) although the draft is now expired. Since that time, RFC 5090 has emerged I notice various differences in the RFC, e.g. the RADIUS server must provide nonces: http://tools.ietf.org/html/rfc5090#section-2.1.5 and may also provide other values. reSIProcate currently generates it's own nonces and passes all the auth parameters to FreeRADIUS for verification. This brings me to some questions: Is anybody already working on migrating FreeRADIUS to the RFC variation of DIGEST support? If so, will older clients stop working? How to handle STUN/TURN? I notice STUN only uses "nonce" and not "qop" or any of the other values, yet RFC 5090 suggests that a RADIUS server can demand that the challenge uses those attributes. I'm also not sure just what value for "method" would be used with STUN. However, it would seem desirable to support STUN/TURN from a single RADIUS server. STUN auth (inherited by TURN) is described here: http://tools.ietf.org/html/rfc5389#section-10.2 This all leaves me feeling that STUN/TURN may need it's now module in FreeRADIUS. Regards, Daniel

1 0

PEAP using different CA?
by Fernando Hammerli 11 Jul '13

11 Jul '13

Hi, Currently we have 1000´s of users self-signed certificates (EAP-TLS), and we´re planning to move our main authentication method to PEAP, but keeping the certificates in use while valid. To avoid the need of installing our CA certificate on every Windows machine, we´ll buy the server certificate from a public CA. Can Freeradius allow me to have both methods at the same time, ie, the PEAP with the public CA and certificate users with our 'self-signed' CA? Thanks in advance, Fernando Hämmerli Pontifícia Universidade Católica do Rio de Janeiro

6 9

Changing proxy server for a specific user
by Angel L. Mateo 11 Jul '13

11 Jul '13

Hello, I have defined a DEFAULT realm proxying requests to a remote radius server (for eduroam). Now I want to create another virtual server using this DEFAULT realm, but proxying requests to another radius server. Is there any way to do this? Maybe modifying some attribute in the request with an entry in the user's file? -- Angel L. Mateo Martínez Sección de Telemática Área de Tecnologías de la Información y las Comunicaciones Aplicadas (ATICA) http://www.um.es/atica Tfo: 868889150 Fax: 868888337

2 1

Reject a specific realm from a client
by David Aldwinckle 11 Jul '13

11 Jul '13

Hello, I have two clients that proxy access requests to me. The realm varies, but the format is always userid(a)realm.whatever Is there a way that I can deny a specific realm when an access request is received from a specific client? I tried adding something to policy.conf but I couldn't get the syntax right: #Prevent secretrealm from logging in off-campus remote_secret_reject if ("%{Realm}" == "secretrealm.ca") && ((Client-Shortname == "proxy-client1") || (Client-Shortname == "proxy-client2"))) { reject } Is there a different way that I should be doing this? Thanks, Dave Dave Aldwinckle Network Support Specialist Information Systems and Technology Phone: (519)-888-4567 ext. 31145 E-Mail: daldwinc(a)uwaterloo.ca<mailto:daldwinc@uwaterloo.ca>

1 0

Indexing multi-valued attrbutes (was RE: 3.0)
by Brian Julin 11 Jul '13

11 Jul '13

Arran Cudbard-Bell wrote: > > Soon. We've gone into official feature freeze. Still finding bugs though, it'd be helpful if people could test. Just to make sure it was understood during the "foreach" fixup patch I sent on github, I mentioned that indexed attribute accesses were broken. None of var[#] var[2] or var[*] work in xlats, unless that's been fixed recently. -- Brian

3 2

freeradius seeing anonymous identity as username
by sebastian buettrich 11 Jul '13

11 Jul '13

hello all, not really a problem, just a question out of interest: i m reconfiguring a freeradius server for eduroam and extended use of rlm_perl for all kinds of fun stuff, and when testing, i notice that all requests containing an anonymous identity @mydomain.org (regardless of where they come from, a real AP or local eapol_test) appear to freeradius as User-Name, i.e. freeradius does not make a difference (?) between anon identity and username is this expected behaviour, the way anonymous identities are implemented, or a sign of some misconfiguration, e.g. someone trying to take care of something that really should just be passed on? it seems to happen before any site configurations are being processed - see debug output and log below. it doesnt do any damage .. just wondering. thanks everybody on the list for sharing, cheers, sebastian ====================================================================== output of radiusd -Xx -------------------------------- Thu Jul 11 10:23:40 2013 : Info: Ready to process requests. rad_recv: Access-Request packet from host 127.0.0.1 port 39508, id=0, length=125 User-Name = "@mydomain.org" NAS-IP-Address = 127.0.0.1 Calling-Station-Id = "CA-FF-EE-00-00" Framed-MTU = 1400 NAS-Port-Type = Wireless-802.11 ... Thu Jul 11 10:23:43 2013 : Info: # Executing section authorize from file /etc/raddb/sites-enabled/default Thu Jul 11 10:23:43 2013 : Info: +- entering group authorize {...} Thu Jul 11 10:23:43 2013 : Info: ++[preprocess] returns ok log file: -------------------------------- Thu Jul 11 10:28:37 2013 : Auth: Login OK: [username/<via Auth-Type = EAP>] (from client local port 0 via TLS tunnel) Thu Jul 11 10:28:37 2013 : Auth: Login OK: [@mydomain.org/<via Auth-Type = EAP>] (from client local port 0 cli CA-FF-EE-00-00)

2 1

Dictionary type problem
by Muhammad Nadeem 11 Jul '13

11 Jul '13

hi everybody I am adding some new attributes to dictionary.motorola. but when i run freeradius , it gives the follwoing error. including dictionary file /usr/local/etc/raddb/dictionary Errors reading dictionary: dict_init: /usr/local/share/freeradius/dictionary.motorola[78]: invalid type "hexdecimal" it clearly indicates that it doesnot support hexadecimal type. Could anyone please tell me , whts the alternative of this. Thanks in advance -- Best Regards Muhammad Nadeem Muhammad Ali Jinnah University

3 3

Update sections
by Arran Cudbard-Bell 10 Jul '13

10 Jul '13

Update sections in 3.0 are considerably more powerful than 2.x.x In addition to being able to override lists on an attribute by attribute basis, e.g: update { request:foo = 'bar' } You can also perform full list copies: update { request: += &reply: } Filtered list copies (all instance of attribute in a list) update { request: += &reply:foo } Reference attributes: update { request:foo += &reply:foo } and perform both types of program execution: update { request:foo := `<my program>` # foo is assigned the output of my program } update { request: := `<my program>` # any VALUE_PAIRs returned by my program are added to the request list } Update sections may now also return fail. (0) update { (0) Executing: "/tmp/foo %{User-Name}" (0) expand: "%{User-Name}" -> 'foo' (0) Program returned code (2): Failed to execute "/tmp/foo": No such file or directory (0) ERROR: Failed to execute "/tmp/foo": No such file or directory (0) } # update = fail In which case subsequent statements are not executed. As with modules failure this can be checked with if (fail). Update sections are no longer atomic as a whole. But individual statements are. That is if a statement fails, all VALUE_PAIRs created up to the point of the failure are discarded. We may be able to restore atomicity of update sections once sets and lists are implemented in 3.1. If you absolutely require it in 3.0 you can wrap each statement in it's own update section and check for failure. -Arran Arran Cudbard-Bell <a.cudbardb(a)freeradius.org> FreeRADIUS Development Team

2 3