March 2011 - Freeradius-Devel - lists.freeradius.org

Unlang and !*
by Brian Candler 20 Mar '11

20 Mar '11

Just a minor point: unlang accepts the syntax if (Foo !* "any") { ... } but it doesn't behave in the same way as !* in a 'users' file. =* and !* are not documented as being available as comparison operators in unlang (only as filtering operators), and there's no particular need for them since you can instead write if (!Foo) { ... } But I think it would be better if it rejected them outright as a syntax error if you tried to use them. It has caused some local confusion recently. Regards, Brian.

2 1

GIT Log for 2011-03-17 00:33 GMT
by aland＠deployingradius.com 16 Mar '11

16 Mar '11

commit 03f1be4fb804522267b3c62b9dc0751efe350662 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 15:17:24 2011 +0100 Allow VSAs of "evs" data type Files changed: src/lib/dict.c | 6 +++--- src/lib/radius.c | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) ====================================================================== commit ed958ed0981ed082bb2095c1f25e1f83f8ece043 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 15:11:16 2011 +0100 Set "flags.is_tlv" correctly Files changed: src/lib/dict.c | 4 +--- src/lib/radius.c | 2 +- 2 files changed, 2 insertions(+), 4 deletions(-) ====================================================================== commit 25f7276880ee3321e97f94b88473fb9adaa2f71e Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 15:09:57 2011 +0100 More descriptive errors in the dictionary parser Files changed: src/lib/dict.c | 49 +++++++++++++++++++++++++++++++++++++++++++------ 1 files changed, 43 insertions(+), 6 deletions(-) ====================================================================== commit d4e2a5d11f022f7ab648e2f656115aa43beda01b Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 13:51:20 2011 +0100 Fix TLV encoding. This seems to be better, as it encodes VSA TLVs correctly. The previous code (vp2attr_tlvs) called do_next_tlv() with the NEXT attribute, which was wrong. Files changed: src/lib/radius.c | 65 +++++++++++++++++++++++++++++++++-------------------- 1 files changed, 40 insertions(+), 25 deletions(-) ====================================================================== commit ae06a5e359c39f3db79bc7df6cc72d70543620d8 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 13:48:02 2011 +0100 Handle invalid formats for extended attributes Files changed: src/lib/print.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) ====================================================================== commit ee170c57224a2ea55813bd765ff9b33dc909f66d Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 11:48:38 2011 +0100 VSA incomplete decode means decode the attribute as "raw" Files changed: src/lib/radius.c | 10 ++++++---- 1 files changed, 6 insertions(+), 4 deletions(-) ====================================================================== commit e8ae8a3b4782a3cef55105308012dfbb36105eea Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 11:26:04 2011 +0100 Expose rad_print_hex, and use it in radsniff Files changed: src/include/libradius.h | 1 + src/lib/radius.c | 4 +--- src/main/radsniff.c | 1 + 3 files changed, 3 insertions(+), 3 deletions(-) ====================================================================== commit 48f452518fa30ddb377eb0efb58c279396df9ee8 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 10:44:31 2011 +0100 Updated for latest features Files changed: man/man5/dictionary.5 | 123 +++++++++++++++++++++++++++++++++---------------- 1 files changed, 83 insertions(+), 40 deletions(-) ====================================================================== commit ca4e6291a37b260b3d0017a0c4e8dea52fd6bdb7 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 08:20:54 2011 +0100 Allow VSAs to be specified as 26.9.1 This makes the OID parsing pretty generic, and it should now work for all attributes. Files changed: src/lib/dict.c | 28 ++++++++++++++++++++++++++++ 1 files changed, 28 insertions(+), 0 deletions(-) ====================================================================== commit e4762aa6d5a7f32dbbea3f23ecd26a2b99717b46 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 16 08:06:45 2011 +0100 Added conflicting dictionary. Alvarion has what, 3-4 different dictionaries for the same vendor space? Files changed: share/dictionary.alvarion.wimax.v2_2 | 34 ++++++++++++++++++++++++++++++++++ 1 files changed, 34 insertions(+), 0 deletions(-) ======================================================================

1 0

GIT Log for 2011-03-16 00:33 GMT
by aland＠deployingradius.com 15 Mar '11

15 Mar '11

commit 7968925c95795767ee89a2b8663b0ebcfe4865fc Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 17:23:22 2011 +0100 Fix compilation errors Files changed: src/main/modcall.c | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-) ====================================================================== commit 65c08416da75f9b1c607bd58f1827f3a639eead1 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 15:22:47 2011 +0100 VP encoding routines now take "const" VP. This means that the API is just a little cleaner.. Files changed: src/include/libradius.h | 11 ++-- src/lib/radius.c | 138 +++++++++++++++++++++++++++-------------------- src/main/radattr.c | 7 +-- 3 files changed, 88 insertions(+), 68 deletions(-) ====================================================================== commit 0df697522d306234d902415c3876b8c4d4be4f23 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 15:22:17 2011 +0100 Id is unsigned for printing Files changed: src/main/event.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) ====================================================================== commit 748472c8f43df54d4bc9851dc2bb4c2db538e263 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 15:19:27 2011 +0100 Added more "const" to functions. This won't make a difference in practice, but it's cleaner. Files changed: src/include/libradius.h | 12 ++++++------ src/lib/filters.c | 2 +- src/lib/misc.c | 2 +- src/lib/print.c | 8 ++++---- 4 files changed, 12 insertions(+), 12 deletions(-) ====================================================================== commit 6ecd67c51cd73f69656bf5adef85021381df1f20 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 14:52:20 2011 +0100 Quiet compiler warning Files changed: src/lib/valuepair.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) ====================================================================== commit 2f1e11482d5f148f76d150e08414f6ce5cb2b91b Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 11:40:55 2011 +0100 Update encoder/decode to use new wimax flags Files changed: src/lib/radius.c | 34 +++++++++++++++++++++------------- 1 files changed, 21 insertions(+), 13 deletions(-) ====================================================================== commit 37ee1cef27e85d3a5603dff03fad7fc989fa94b2 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 11:31:42 2011 +0100 Add a "wimax" format flag to the attributes. This is in preparation for horrible Alvarion hacks Files changed: src/include/libradius.h | 1 + src/lib/dict.c | 9 +++++++++ 2 files changed, 10 insertions(+), 0 deletions(-) ====================================================================== commit 390f6efd70ad2a4172bc63d7373a4e0064280d6c Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 11:27:47 2011 +0100 Fix build error Files changed: src/modules/rlm_sqlcounter/rlm_sqlcounter.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) ====================================================================== commit 0b2727f7e85e3f904c6a2a8c3a291656e604462f Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 09:45:11 2011 +0100 If select() returns an error, show it to the user. Helps to address bug #149 However, the underlying issue still isn't fixed. It isn't clear why select() returns an error. An alternate solution might be to simply re-do all of the FD_SET stuff, in the hope that the cached version was wrong. Files changed: src/lib/event.c | 4 +++- src/main/radiusd.c | 12 +++++++++--- 2 files changed, 12 insertions(+), 4 deletions(-) ====================================================================== commit 3161e60705e553789d421c70b7f3e4f3b6c27b95 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 15 09:34:54 2011 +0100 Note bad configuration The sqlcounter module does it's own expansion. This doesn't seem to work. Files changed: raddb/modules/sqlcounter_expire_on_login | 15 ++++++++++++++- src/modules/rlm_sqlcounter/rlm_sqlcounter.c | 2 ++ 2 files changed, 16 insertions(+), 1 deletions(-) ====================================================================== commit 497465d96daa19805d7926cd0732144f321d6723 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Mon Mar 14 14:15:21 2011 +0100 Updated command list radmin -e "help -r" | perl -ne 's/^(.*) - /\n/;$foo = $1; $foo =~ s/\ /\\ /g;print ".IP ", $foo;print;' > foo Files changed: man/man8/radmin.8 | 26 ++++++++++++++++++++++---- 1 files changed, 22 insertions(+), 4 deletions(-) ======================================================================

1 0

Re: Possible bug in configurable failover
by Brian Candler 15 Mar '11

15 Mar '11

Alan DeKok wrote: > The "update" section should behave identically to the "if". It doesn't at the moment. Version 1 policy.conf: testing_module { if (1) { update reply { Reply-Message += "Foo" } } if (1) { update reply { Reply-Message += "Bar" } } ok } Results: $ bin/radtest steve badpass localhost 1 testing123 Sending Access-Request of id 69 to 127.0.0.1 port 1812 User-Name = "steve" User-Password = "badpass" NAS-IP-Address = 127.0.0.1 NAS-Port = 1 rad_recv: Access-Reject packet from host 127.0.0.1 port 1812, id=69, length=25 Reply-Message = "Foo" $ Version 2 policy.conf: testing_module { #if (1) { update reply { Reply-Message += "Foo" } #} #if (1) { update reply { Reply-Message += "Bar" } #} ok } Results: $ bin/radtest steve badpass localhost 1 testing123 Sending Access-Request of id 33 to 127.0.0.1 port 1812 User-Name = "steve" User-Password = "badpass" NAS-IP-Address = 127.0.0.1 NAS-Port = 1 rad_recv: Access-Accept packet from host 127.0.0.1 port 1812, id=33, length=81 Service-Type = Framed-User Framed-Protocol = PPP Framed-IP-Address = 172.16.3.33 Framed-IP-Netmask = 255.255.255.0 Framed-Routing = Broadcast-Listen Filter-Id = "std.ppp" Framed-MTU = 1500 Framed-Compression = Van-Jacobson-TCP-IP Reply-Message = "Foo" Reply-Message = "Bar" $ So: without the 'ifs', the three sections are run sequentially (update, update, ok). With the 'ifs', only the first section is run. Authenticate section has: Auth-Type PAP { pap { ok = return reject = 1 } testing_module } Regards, Brian.

2 8

Possible bug in logging duplicate requests
by Alexandr Kovalenko 15 Mar '11

15 Mar '11

Hello! I've found that when duplicate request found it is logged with (probably) wrong printf pattern: src/main/event.c line 2574: ------------------ radlog(L_ERR, "Discarding duplicate request from " "client %s port %d - ID: %d due to unfinished request %u", client->shortname, request->packet->src_port,request->packet->id, request->number); break; ----------------- Due to 'ID: %d' I'm getting negative IDs in log: Tue Mar 15 11:46:57 2011 : Error: Discarding duplicate request from client dhcp port 67 - ID: -2021971096 due to unfinished request 977 I think it should be changed to '%u'. -- Alexandr Kovalenko http://uafug.org.ua/

2 1

Possible bug in configurable failover
by Brian Candler 15 Mar '11

15 Mar '11

I think I've found a bug in configurable failover - or at least, it doesn't correspond to any behaviour I can see documented, but I thought I'd raise it here before making a ticket. To replicate: start with stock freeradius 2.1.x, uncomment the "steve" entry from the users file, and apply the following config changes: --- etc/raddb/sites-available/default.orig 2011-03-13 19:48:20.584961000 +0000 +++ etc/raddb/sites-available/default 2011-03-13 19:47:09.244961001 +0000 @@ -237,7 +237,11 @@ # in the 'authorize' section supplies a password. The # password can be clear-text, or encrypted. Auth-Type PAP { - pap + pap { + ok = return + reject = 1 + } + testing_module } # --- etc/raddb/policy.conf.orig 2011-03-13 19:48:26.154961000 +0000 +++ etc/raddb/policy.conf 2011-03-13 19:49:45.434961000 +0000 @@ -194,4 +194,18 @@ noop } } + + testing_module { + if (1) { + update reply { + Reply-Message += "Foo" + } + } + if (1) { + update reply { + Reply-Message += "Bar" + } + ok + } + } } The idea is using a module to turn an access reject into an access reject under certain circumstances. It just happens to have two separate 'if' statements (and this is what triggers the bug) Everything is fine if you give the correct password for steve. If you give the wrong password, something strange happens: $ bin/radtest steve badpass localhost 1 testing123 Sending Access-Request of id 221 to 127.0.0.1 port 1812 User-Name = "steve" User-Password = "badpass" NAS-IP-Address = 127.0.0.1 NAS-Port = 1 rad_recv: Access-Reject packet from host 127.0.0.1 port 1812, id=221, length=25 Reply-Message = "Foo" You can see that only the first 'if' statement has been executed, and then it has dropped out of the module entirely, keeping the reject status. As a workaround, you can insert a 'noop': --- etc/raddb/policy.conf.orig 2011-03-13 19:48:26.154961000 +0000 +++ etc/raddb/policy.conf 2011-03-13 19:57:42.294961000 +0000 @@ -194,4 +194,19 @@ noop } } + + testing_module { + if (1) { + update reply { + Reply-Message += "Foo" + } + noop + } + if (1) { + update reply { + Reply-Message += "Bar" + } + ok + } + } } And then it behaves how I would have expected in the first place, without the noop: $ bin/radtest steve badpass localhost 1 testing123 Sending Access-Request of id 159 to 127.0.0.1 port 1812 User-Name = "steve" User-Password = "badpass" NAS-IP-Address = 127.0.0.1 NAS-Port = 1 rad_recv: Access-Accept packet from host 127.0.0.1 port 1812, id=159, length=81 Service-Type = Framed-User Framed-Protocol = PPP Framed-IP-Address = 172.16.3.33 Framed-IP-Netmask = 255.255.255.0 Framed-Routing = Broadcast-Listen Filter-Id = "std.ppp" Framed-MTU = 1500 Framed-Compression = Van-Jacobson-TCP-IP Reply-Message = "Foo" Reply-Message = "Bar" Any thoughts as to what is going on here? Thanks, Brian.

2 3

GIT Log for 2011-03-15 00:33 GMT
by aland＠deployingradius.com 14 Mar '11

14 Mar '11

commit c3b5aa29a3c7d1afaca8a4d054e563246fa7fe26 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Mon Mar 14 14:01:28 2011 +0100 Check pre-condictions "inject file" requires "inject to" and "inject from" Files changed: src/main/command.c | 10 ++++++++++ 1 files changed, 10 insertions(+), 0 deletions(-) ====================================================================== commit 721c32b3b3e1ee2e01c8b1ac7b5e81566ce3480c Author: Alan T. DeKok <aland(a)freeradius.org> Date: Mon Mar 14 12:04:56 2011 +0100 If the source address is INADDR_ANY, don't use udpfromto It breaks various systems. Closes bug #148 Files changed: src/lib/radius.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) ====================================================================== commit 767ac8bafc62350df0f47ce5e606636e1593b3c8 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Mon Mar 14 11:59:53 2011 +0100 More sanity checks on decoding extended attrs Files changed: src/lib/radius.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) ====================================================================== commit 90ef1ab273bba4fb29501b1b8582eade8072e313 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Mon Mar 14 11:56:32 2011 +0100 Enforce numerical limits on "extended" attributes. They can only be in the range 241..255. This ensures that the internal code which packs the attribute number into the high octet of the vendor ID will work. If it allowed "extended" attrs to be 1..30 (or so), other things might potentially screw up internally Files changed: src/lib/dict.c | 11 ++++++++++- 1 files changed, 10 insertions(+), 1 deletions(-) ====================================================================== commit 38715dfceba904c0569d1800186e0c967c188c8a Author: Alan T. DeKok <aland(a)freeradius.org> Date: Mon Mar 14 10:25:45 2011 +0100 Moved USR/ascend illegal dictionaries back to share/ raddb/dictionary is *not* changed on an upgrade, so removing those dictionaries from share/dictionary means that they might be surprisingly removed from the local configuration, and break things. That would be bad. Files changed: raddb/dictionary.in | 17 ----------------- share/dictionary | 9 +++++++-- 2 files changed, 7 insertions(+), 19 deletions(-) ======================================================================

1 0

my first freeradius module
by Herbert Fischer 14 Mar '11

14 Mar '11

Hi, I'm kinda new to Freeradius and at the same time I'm learning it, I'm seeing I gonna need to develop a new module for it. This module will do OTP two way authentication. It will extract part of the password (ex.: latest 6 digits) to verify and the remaining "password" will be returned to Freeradius to test against another module (LDAP for example). In this module I need to store user information on some place, including usernames, keys, secrets, last OTP, offset, etc. I have some options: * Use some file like the "users" file from freeradius or a tabulated file. Read this file in memory and update user information in memory and in file when needed, so if Freeradius is restarted, or the machines crashes, the updated information is not lost. However I don't know exactly if I'm going to have problems with the "module needs to update file" part. Don't know how Freeradius's internal API/libs work with files. I'm trying to understand rlm_files.c but this module only reads the "users" file. * Use MySQL or Redis to store this information. As they support atomic operations I should have no problem using them. However I gonna need to link the module against some external lib and I fear this may bring Freeradius some problems, and I'll have another service to monitor, manage, etc. What do you suggest? Is there any other way to do this two way authentication without needing to develop a module for it? thanks in advance, Herbert

3 6

GIT Log for 2011-03-14 00:33 GMT
by aland＠deployingradius.com 13 Mar '11

13 Mar '11

commit 1b839fb77c533edea1ee99673a20e3d2949a1470 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Sun Mar 13 18:50:54 2011 +0100 Revert "Fix typos and regenerate for 2.1.11" This reverts commit 2d0c5fa94938f727b4f37fb7399b777437171eeb. Files changed: configure | 200 ++++++++++++++++++++++++++++++++++++++++++++++++++++--------- 1 files changed, 172 insertions(+), 28 deletions(-) ====================================================================== commit 842c4d887e44832159f34641127064d9ee75b05e Author: Alan T. DeKok <aland(a)freeradius.org> Date: Sun Mar 13 10:12:11 2011 +0100 Complain about duplicate virtual servers, too Files changed: src/main/modules.c | 44 +++++++++++++++++++++++++++++++++----------- 1 files changed, 33 insertions(+), 11 deletions(-) ====================================================================== commit fc393073000c17ae122ff0cbdc1c831eb3642d4e Author: Alan T. DeKok <aland(a)freeradius.org> Date: Sun Mar 13 10:02:33 2011 +0100 Look for duplicate module definitions. Doing an O(N^2) search over the modules{} section. If the same module is defined twice, print an error detailing *both* locations where it occurs, and exit. This helps prevent broken configurations by disallowing the server from loading two conflicting module definitions Files changed: src/main/modules.c | 39 ++++++++++++++++++++++++++++++++++++++- 1 files changed, 38 insertions(+), 1 deletions(-) ====================================================================== commit fbf4bb8fa1e8c1238e317fa0b6a2e2a7c08b18c2 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Sun Mar 13 10:02:06 2011 +0100 Added cf_section_find_name2() API This lets us find the next section by name1 and name2 Files changed: src/include/conffile.h | 2 ++ src/main/conffile.c | 27 +++++++++++++++++++++++++++ 2 files changed, 29 insertions(+), 0 deletions(-) ====================================================================== commit d228cd073ac13196980a813df8f2ad0cb5e6dc0d Author: Alan T. DeKok <aland(a)freeradius.org> Date: Fri Mar 11 10:49:43 2011 +0100 Added dictionary from Moonshot project Files changed: share/dictionary | 2 ++ share/dictionary.ukerna | 18 ++++++++++++++++++ 2 files changed, 20 insertions(+), 0 deletions(-) ====================================================================== commit df9dfae2c1949514894e4e36a7172402714406e5 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Sun Mar 13 10:14:30 2011 +0100 Added duplicate names for simplicity byte == uint8 short == uint16 integer == uint32 signed == int32 Files changed: src/lib/dict.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) ======================================================================

1 0

Compilation broken in v2.1.x branch?
by Brian Candler 13 Mar '11

13 Mar '11

I hadn't done a 'git pull' and recompile since 1st February. Now when I try it again, the build fails on my machine (Ubuntu 10.10 x86_64) $ make distclean && ./configure --prefix=/v/build/fr && make && make install ... gcc -g -O2 -D_REENTRANT -D_POSIX_PTHREAD_SEMANTICS -Wall -D_GNU_SOURCE -g -Wshadow -Wpointer-arith -Wcast-qual -Wcast-align -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wnested-externs -W -Wredundant-decls -Wundef -I/u/home/brian/git/freeradius-server/src -DHOSTINFO=\"x86_64-unknown-linux-gnu\" -DRADIUSD_VERSION=\"2.1.11\" -DOPENSSL_NO_KRB5 -c detail.c -o detail.o >/dev/null 2>&1 /u/home/brian/git/freeradius-server/libtool --mode=link gcc -export-dynamic -dlopen self \ -o radiusd acct.lo auth.lo client.lo conffile.lo crypt.lo exec.lo files.lo listen.lo log.lo mainconfig.lo modules.lo modcall.lo radiusd.lo stats.lo soh.lo session.lo threads.lo util.lo valuepair.lo version.lo xlat.lo event.lo realms.lo evaluate.lo vmps.lo detail.lo \ /u/home/brian/git/freeradius-server/src/lib/libfreeradius-radius.la -lnsl -lresolv -lpthread \ -lcrypt -lltdl -lcrypto -lssl -lcrypto rm -f .libs/radiusd.nm .libs/radiusd.nmS .libs/radiusd.nmT creating .libs/radiusdS.c (cd .libs && gcc -g -O2 -c -fno-builtin "radiusdS.c") rm -f .libs/radiusdS.c .libs/radiusd.nm .libs/radiusd.nmS .libs/radiusd.nmT gcc .libs/radiusdS.o -o .libs/radiusd .libs/acct.o .libs/auth.o .libs/client.o .libs/conffile.o .libs/crypt.o .libs/exec.o .libs/files.o .libs/listen.o .libs/log.o .libs/mainconfig.o .libs/modules.o .libs/modcall.o .libs/radiusd.o .libs/stats.o .libs/soh.o .libs/session.o .libs/threads.o .libs/util.o .libs/valuepair.o .libs/version.o .libs/xlat.o .libs/event.o .libs/realms.o .libs/evaluate.o .libs/vmps.o .libs/detail.o -Wl,--export-dynamic /u/home/brian/git/freeradius-server/src/lib/.libs/libfreeradius-radius.so -lnsl -lresolv -lpthread -lcrypt /usr/lib/libltdl.so -lssl -lcrypto -ldl -Wl,--rpath -Wl,/v/build/fr/lib .libs/modules.o: In function `setup_modules': /v/git/freeradius-server/src/main/modules.c:1395: undefined reference to `lt__PROGRAM__LTX_preloaded_symbols' collect2: ld returned 1 exit status make[4]: *** [radiusd] Error 1 make[4]: Leaving directory `/v/git/freeradius-server/src/main' make[3]: *** [main] Error 2 make[3]: Leaving directory `/v/git/freeradius-server/src' make[2]: *** [all] Error 2 make[2]: Leaving directory `/v/git/freeradius-server/src' make[1]: *** [src] Error 2 make[1]: Leaving directory `/v/git/freeradius-server' make: *** [all] Error 2 Doing a git bisect from 5853570817fc2a1b5e972fbafe10568813e426ac suggests the problem to be introduced here: 2d0c5fa94938f727b4f37fb7399b777437171eeb is the first bad commit commit 2d0c5fa94938f727b4f37fb7399b777437171eeb Author: Alan T. DeKok <aland(a)freeradius.org> Date: Mon Feb 14 15:03:48 2011 +0100 Fix typos and regenerate for 2.1.11 :100755 100755 cd0356c27a219e968bf3cecaef08ca04f74eea64 d48179f54bd576f474168325dc0aaac2d9c4dfe0 M configure And to confirm this manually: - a checkout of 995b62f8 builds successfully - bc9435fd builds successfully - 2d0c5fa9 fails as above The relevant software versions are: ii autoconf 2.67-2ubuntu1 automatic configure script builder ii automake 1:1.11.1-1 A tool for generating GNU Standards-compliant Makefiles ii autotools-dev 20100122.1 Update infrastructure for config.{guess,sub} files ii libltdl-dev 2.2.6b-2ubuntu1 A system independent dlopen wrapper for GNU libtool ii libltdl7 2.2.6b-2ubuntu1 A system independent dlopen wrapper for GNU libtool ii libtool 2.2.6b-2ubuntu1 Generic library support script I'm no expert in autoconf black magic, but it seems to me that the "fix typos" actually introduced one (it should be libltdl not libtldl) Regards, Brian.

2 2