January 2006 - Freeradius-Devel - lists.freeradius.org

compile & install on FC4 x86_64
by FreeRadius Devel 15 May '06

15 May '06

hello, I am trying to install freeradius cvs version downloaded on 2006-01-14 on Fedora CORE 4 installed on a dual AMD Opteron system. fedora is updated as of 2006-01-14. I have have found two problems and I do not know if they are fedora, x86_64 or freeradius related. 1. after: # ./configure --with-large-files --with-udpfromto # make I get the error: gcc: /usr/lib/libsnmp.so: No such file or directory gmake[4]: *** [radiusd] Error 1 gmake[4]: Leaving directory `.../freeradius-060114/src/main' there is no file "/usr/lib/libsnmp.so" but there is a file "/usr/lib64/libsnmp.so" is the configure script suppose to find the 32bit version or the 64bit version ? after a link from /usr/lib64/libsnmp.so to /usr/lib/libsnmp.so freeradius compiles and installs 2. after running "# radiusd -X" I get the following error: Starting - reading configuration files ... read_config_files: reading dictionary Errors reading dictionary: dict_init: /usr/local/share/freeradius/dictionary.ascend[1228]: dict_addvalue: Duplicate value name Route-IPX-No for attribute X-Ascend-Route-IPX Segmentation fault identical configuration & installation works on FC4/i386 in file "src/lib/dict.c", function "DICT_VALUE *dict_valbyname(int attr, const char *name)" line "hash = lrad_hash_update(&attr, sizeof(&attr), hash);" I think that "sizeof(&attr)" should be "sizeof(attr)" because sizeof(int) != sizeof(&int) on 64bit. is there a compile option hat would instruct the compiler to consider "int" as being a 64bit integer ? thanks, Razvan Radu

4 3

Release 1.1.1 TODO
by Nicolas Baradakis 15 Mar '06

15 Mar '06

I don't know if it's too soon to plan a release date for 1.1.1, as 1.1.0 has some annoying known problems: - an error on x86_64 machines when reading dictionnaries. - compilation fails on FreeBSD because of rlm_otp. - rlm_eap_peap and rlm_eap_ttls have libtool problems on Debian. The following list is surely very incomplete, but I tried to put together the things we could do before the next release. - pull eap fixes from CVS head to address Debian's issues. I found many changes not in branch 1.1 but I've no idea whether they are safe or not: - moved TLS code from rlm_eap_tls to libeap. - removed record_* functions from global visibility. - remove the EAP submodule locking. - change autoconf tests for OpenSSL. - put the eap sessions into a tree, so that looking them up is very fast, and no longer O(n) in the number of sessions. - if the callback fails, do eap_fail() to get an EAP-Failure message and return reject. - use new hex2bin function to be more forgiving for NT-Passwords - make "use_tunneled_reply" work properly for PEAP - fix bug #314 about rlm_otp: the module could have more tests in configure.in so it is silently disabled instead of giving compilation errors. - net-snmp patches. (and move #includes from radius_snmp.h to smux.h) - maybe undelete the database schema for MSSQL server? We could put the file under src/modules/rlm_sql/drivers/rlm_sql_unixodbc for example. - maybe back-port the changes in the makefiles in CVS head? It would fix some warnings and make the autoconf/libtool upgrade easier in the future. -- Nicolas Baradakis

8 27

RE: using stored procedures with freeradius (Urgent help request)
by Murat Mığdısoğlu 18 Feb '06

18 Feb '06

I found it, If u return some variables from the accounting sp with some kind of select, you can encounter this problem. -----Original Message----- From: freeradius-devel-bounces+murat.migdisoglu=superonline.net(a)lists.freeradius.org [mailto:freeradius-devel-bounces+murat.migdisoglu=superonline.net@lists.freeradius.org] On Behalf Of Murat Mığdısoğlu Sent: Saturday, January 28, 2006 3:29 PM To: freeradius-devel(a)lists.freeradius.org Subject: using stored procedures with freeradius (Urgent help request) Hi all, I'm using freeradius with sybase using freetds and unixodbc. For some purposes, i had to use stored procedures and changed sql statements in sql.con to procedure calls like "EXEC -----". I have to question at this point 1) has anyone used this method before?(in the archie, i saw some mails about sp, so i assume freeradius support it) 2) Examining my logs, i found that some sockets getting 'Invalid cursor State' error from unix-odbc driver in some cases and they don't work anymore. What it can be? Bu e-posta mesaji ve ekleri sadece gonderildigi kisi veya kuruma ozeldir. Eger dogru kisiye ulasmadigini dusunuyorsaniz, bu mesajin yonlendirilmesi, kopyalanmasi veya herhangi bir sekilde kullanilmasi yasaktir.Mesaj iceriginde bulunan fikir ve yorumlar, SUPERONLINE'a degil sadece gondericiye aittir. Bu mesaj bilinen tum viruslere karsi test edilmistir. This e-mail and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you are not the intended recipient you are hereby notified that any dissemination, copying or use of the information is prohibited. The opinions expressed in this message belong to sender alone. There is no implied endorsement by SUPERONLINE.This e-mail has been scanned for all known computer viruses. - List info/subscribe/unsubscribe? See http://www.freeradius.org/list/devel.html Bu e-posta mesaji ve ekleri sadece gonderildigi kisi veya kuruma ozeldir. Eger dogru kisiye ulasmadigini dusunuyorsaniz, bu mesajin yonlendirilmesi, kopyalanmasi veya herhangi bir sekilde kullanilmasi yasaktir.Mesaj iceriginde bulunan fikir ve yorumlar, SUPERONLINE'a degil sadece gondericiye aittir. Bu mesaj bilinen tum viruslere karsi test edilmistir. This e-mail and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you are not the intended recipient you are hereby notified that any dissemination, copying or use of the information is prohibited. The opinions expressed in this message belong to sender alone. There is no implied endorsement by SUPERONLINE.This e-mail has been scanned for all known computer viruses.

2 1

[PATCH 1/10] rlm_python: 01---sanitize-rlm-python-t.patch
by Paul P Komkoff Jr 09 Feb '06

09 Feb '06

- Sanitize rlm_python_t - Remove typedef - Introduce global PyObject radiusd_module (anyway we have only one interpreter) --- commit 20b0202331f6bc26c46d3c91caf3c209a5958e01 tree 11c4bc223c448c60c4adce48e3fa56840be53e6d parent fef24dfbd7872636bc466bb28637578b1f8604ca author <stingray(a)boxster64.stingr.net> Sun, 29 Jan 2006 14:44:46 +0300 committer <stingray(a)boxster64.stingr.net> Sun, 29 Jan 2006 14:44:46 +0300 src/modules/rlm_python/rlm_python.c | 128 ++++++++++++++++------------------- 1 files changed, 60 insertions(+), 68 deletions(-) diff --git a/src/modules/rlm_python/rlm_python.c b/src/modules/rlm_python/rlm_python.c --- a/src/modules/rlm_python/rlm_python.c +++ b/src/modules/rlm_python/rlm_python.c @@ -42,55 +42,43 @@ static const char rcsid[] = "$Id: rlm_py * a lot cleaner to do so, and a pointer to the structure can * be used as the instance handle. */ -typedef struct rlm_python_t { - /* Config section */ - /* Names of modules */ - char - *mod_instantiate, - *mod_authorize, - *mod_authenticate, - *mod_preacct, - *mod_accounting, - *mod_checksimul, - *mod_detach, - - /* Names of functions */ - *func_instantiate, - *func_authorize, - *func_authenticate, - *func_preacct, - *func_accounting, - *func_checksimul, - *func_detach; - - - /* End Config section */ - - - /* Python objects for modules */ - PyObject - *pModule_builtin, - *pModule_instantiate, - *pModule_authorize, - *pModule_authenticate, - *pModule_preacct, - *pModule_accounting, - *pModule_checksimul, - *pModule_detach, - - - /* Functions */ - - *pFunc_instantiate, - *pFunc_authorize, - *pFunc_authenticate, - *pFunc_preacct, - *pFunc_accounting, - *pFunc_checksimul, - *pFunc_detach; +struct rlm_python_t { + char *mod_instantiate; + char *mod_authorize; + char *mod_authenticate; + char *mod_preacct; + char *mod_accounting; + char *mod_checksimul; + char *mod_detach; + + /* Names of functions */ + char *func_instantiate; + char *func_authorize; + char *func_authenticate; + char *func_preacct; + char *func_accounting; + char *func_checksimul; + char *func_detach; + + PyObject *pModule_instantiate; + PyObject *pModule_authorize; + PyObject *pModule_authenticate; + PyObject *pModule_preacct; + PyObject *pModule_accounting; + PyObject *pModule_checksimul; + PyObject *pModule_detach; + + /* Functions */ + PyObject *pFunc_instantiate; + PyObject *pFunc_authorize; + PyObject *pFunc_authenticate; + PyObject *pFunc_preacct; + PyObject *pFunc_accounting; + PyObject *pFunc_checksimul; + PyObject *pFunc_detach; +}; -} rlm_python_t; /* * A mapping of configuration file names to internal variables. @@ -103,44 +91,49 @@ typedef struct rlm_python_t { */ static CONF_PARSER module_config[] = { { "mod_instantiate", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, mod_instantiate), NULL, NULL}, + offsetof(struct rlm_python_t, mod_instantiate), NULL, NULL}, { "func_instantiate", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, func_instantiate), NULL, NULL}, + offsetof(struct rlm_python_t, func_instantiate), NULL, NULL}, { "mod_authorize", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, mod_authorize), NULL, NULL}, + offsetof(struct rlm_python_t, mod_authorize), NULL, NULL}, { "func_authorize", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, func_authorize), NULL, NULL}, + offsetof(struct rlm_python_t, func_authorize), NULL, NULL}, { "mod_authenticate", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, mod_authenticate), NULL, NULL}, + offsetof(struct rlm_python_t, mod_authenticate), NULL, NULL}, { "func_authenticate", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, func_authenticate), NULL, NULL}, + offsetof(struct rlm_python_t, func_authenticate), NULL, NULL}, { "mod_preacct", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, mod_preacct), NULL, NULL}, + offsetof(struct rlm_python_t, mod_preacct), NULL, NULL}, { "func_preacct", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, func_preacct), NULL, NULL}, + offsetof(struct rlm_python_t, func_preacct), NULL, NULL}, { "mod_accounting", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, mod_accounting), NULL, NULL}, + offsetof(struct rlm_python_t, mod_accounting), NULL, NULL}, { "func_accounting", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, func_accounting), NULL, NULL}, + offsetof(struct rlm_python_t, func_accounting), NULL, NULL}, { "mod_checksimul", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, mod_checksimul), NULL, NULL}, + offsetof(struct rlm_python_t, mod_checksimul), NULL, NULL}, { "func_checksimul", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, func_checksimul), NULL, NULL}, + offsetof(struct rlm_python_t, func_checksimul), NULL, NULL}, { "mod_detach", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, mod_detach), NULL, NULL}, + offsetof(struct rlm_python_t, mod_detach), NULL, NULL}, { "func_detach", PW_TYPE_STRING_PTR, - offsetof(rlm_python_t, func_detach), NULL, NULL}, + offsetof(struct rlm_python_t, func_detach), NULL, NULL}, { NULL, -1, 0, NULL, NULL } /* end the list */ }; + +/* Let assume that radiusd module is only one since we have only one intepreter */ + +static PyObject *radiusd_module = NULL; + /* * radiusd Python functions */ @@ -570,8 +563,7 @@ static int load_python_function(const ch */ static int python_instantiate(CONF_SECTION *conf, void **instance) { - rlm_python_t *data; - PyObject *module; + struct rlm_python_t *data; int idx; /* @@ -598,7 +590,7 @@ static int python_instantiate(CONF_SECTI */ /* Code */ - if ((module = data->pModule_builtin = + if ((radiusd_module = Py_InitModule3("radiusd", radiusd_methods, "FreeRADIUS Module.")) == NULL) { @@ -611,7 +603,7 @@ static int python_instantiate(CONF_SECTI * Load constants into module */ for (idx=0; constants[idx].name; idx++) - if ((PyModule_AddIntConstant(module, constants[idx].name, constants[idx].value)) == -1) { + if ((PyModule_AddIntConstant(radiusd_module, constants[idx].name, constants[idx].value)) == -1) { radlog(L_ERR, "Python AddIntConstant failed"); } @@ -717,15 +709,15 @@ static int python_detach(void *instance) /* Default return value is failure */ return_value = -1; - if (((rlm_python_t *)instance)->pFunc_detach && - PyCallable_Check(((rlm_python_t *)instance)->pFunc_detach)) { + if (((struct rlm_python_t *)instance)->pFunc_detach && + PyCallable_Check(((struct rlm_python_t *)instance)->pFunc_detach)) { PyObject *pArgs, *pValue; /* call the function with an empty tuple */ pArgs = PyTuple_New(0); - pValue = PyObject_CallObject(((rlm_python_t *)instance)->pFunc_detach, + pValue = PyObject_CallObject(((struct rlm_python_t *)instance)->pFunc_detach, pArgs); if (pValue == NULL) { -- Paul P 'Stingray' Komkoff Jr // http://stingr.net/key <- my pgp key This message represents the official view of the voices in my head

5 8

Repeated usage of regex results like %{x}
by Stephan Jaeger 08 Feb '06

08 Feb '06

Hello, request_data_get() deletes the data from the request after retrieving it, therefore it is not possible to use a regex value e.g. %{1} in rlm_policy twice like: request = { User-Name := "%{1}" Nas-Identifier := "%{1}" } What would be the best way to change this or work around it? Regards Stephan -- Stephan Jaeger <stephan.jaeger(a)ewetel.de> EWE TEL GmbH

2 3

Release 1.1.0 this week?
by Alan DeKok 01 Feb '06

01 Feb '06

I was planning on releasing it Monday, but now Tuesday or Wednesday is looking better for me. Comments? Also, here's a list of what could be in later releases 1.1.1 - libeap fixes - netsnmp fixes 1.1.2 - upgraded libtool & autoconf?? Once that's done, the CVS head is starting to look a little anemic... Alan DeKok.

5 8

rfc2865 and RADIUS-LDAPv3.schema inconsistencies?
by Markus Krause 31 Jan '06

31 Jan '06

hi developers! i had a closer look at RADIUS-LDAPv3.schema (rev.1.9.6.1) and compared it with rfc2865 and there seem to be some "inconsistencies": some attributes which may contain zero or more values according to rfc2865 (http://rfc.net/rfc2865.html) are defined as SINGLE-VALUE in the schema file. as far as i understand the rfc the following values (the numbers in brackets refer to the section nr. and page) may contain more values: Filter-Id <-> radiusFilterId (5.11, p.36) Framed-AppleTalk-Network <-> radiusFramedAppleTalkNetwork (5.38, p.58) Framed-Compresion <-> radiusFramedCompression (5.13, p.37) Login-IP-Host <-> radiusLoginIPHost (5.15, p.38) the attribute NAS-IP-Address (radiusNASIpAddress) is described (in rfc) to be allowed to be present zero or one time (see 5.4, p.29, and 5.44: table of attributes, p.63), but is not defined as SINGLE-VALUE in the schema file. i am not using these values right now so i don't know if this differences can cause problems, i just came across this by testing nicolas' patch for bug 261 some time ago and finally had some time and mood to read the rfc. of course i also do not know if these differences are intended, i am just wondering about them! ;-) with best regards, markus -- Markus Krause email: krause(a)biochem.mpg.de Computing Center Tel.: 089 - 89 40 85 99 Group Lottspeich / Proteomics Fax.: 089 - 89 40 85 98 --------------------------------------------------------------------- This message was sent using https://webmail.biochem.mpg.de If you encounter any problems please report to rz-linux(a)biochem.mpg.de

4 7

Automatic report from sources (radiusd) between 30.01.2006 - 31.01.2006 GMT
by Automatic cvs log generator 31 Jan '06

31 Jan '06

CVS log entries from 30.01.2006 (Mon) 09:00:00 - 31.01.2006 (Tue) 09:00:01 GMT ===================================================== Summary by authors ===================================================== Author: aland File: radiusd/src/modules/rlm_pap/rlm_pap.c; Revisions: 1.36 File: radiusd/src/lib/hash.c; Revisions: 1.10 ===================================================== Log entries ===================================================== Description: Don't increment num_elements too often Modified files: File: radiusd/src/lib/hash.c; Revision: 1.10; Date: 2006/01/30 22:06:10; Author: aland; Lines: (+3 -4) ------------------------------- Description: Update warning message Modified files: File: radiusd/src/modules/rlm_pap/rlm_pap.c; Revision: 1.36; Date: 2006/01/30 19:16:10; Author: aland; Lines: (+3 -3) ===================================================== Summary of modified files ===================================================== File: radiusd/src/lib/hash.c Revisions: 1.10 Authors: aland (+3 -4) ------------------------------- File: radiusd/src/modules/rlm_pap/rlm_pap.c Revisions: 1.36 Authors: aland (+3 -3) -- Automatic cron job from /web/pages/us.freeradius.org/bin/new_makelog.pl

1 0

rlm_ippool failed to get Pool-Name attribute on solaris 10
by German Gumenyuk 31 Jan '06

31 Jan '06

Hello Alan, seems i have used wrong Client-IP-Address value in DEFAULT entry. thank you for help. German Gumenyuk --------------------------------- Yahoo! Autos. Looking for a sweet ride? Get pricing, reviews, & more on new and used cars.

1 0

rlm_ippool failed to get Pool-Name attribute on solaris 10
by German Gumenyuk 31 Jan '06

31 Jan '06

Hello Alan, thank you for answer. in debug log it is matched correct user. this user has Fall-Through = yes in users file Pool-Name entries are defined in following DEFAULT entries (there is part of users file in previous mail) But pairfind() returns NULL for PW_POOL_NAME in rlm_ippool module: ------ Processing the authorize section of radiusd.conf modcall: entering group authorize for request 0 modcall[authorize]: module "preprocess" returns ok for request 0 users: Matched entry userms4 at line 1 modcall[authorize]: module "files" returns ok for request 0 modcall: group authorize returns ok for request 0 rad_check_password: Found Auth-Type Local auth: type Local auth: user supplied CHAP-Password matches local User-Password Processing the post-auth section of radiusd.conf modcall: entering group post-auth for request 0 rlm_ippool: Could not find Pool-Name attribute. modcall[post-auth]: module "p2_2" returns noop for request 0 rlm_ippool: Could not find Pool-Name attribute. ... ----- German Gumenyuk --------------------------------- What are the most popular cars? Find out at Yahoo! Autos

1 0