[Bug 1850927] Re: Configuration comment suggests non-default /tmp/radius configuration which doesn't exist by default

2022-03-04 Thread Alan DeKok
> Oct 30 14:14:56 radius freeradius[5524]: TLS section "tls" missing, trying to use legacy configuration i.e. you edited the default "eap" module configuration to *remove* the "tls" configuration. Then tried to use the "eap-tls" method. And wondered why it didn't work. Here's the solution:

[Bug 1962046] Re: DEP8: test more auth algorithms

2022-02-24 Thread Alan DeKok
Fix pushed https://github.com/FreeRADIUS/freeradius- server/commit/beb1351d682ab6f1170cc60ce7ca05f597e5d7d6 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1962046 Title: DEP8: test more auth

[Bug 1962046] Re: DEP8: test more auth algorithms

2022-02-23 Thread Alan DeKok
3.0.26 isn't released. But if the git "head" of v3.0.x crashes, that's bad. Perhaps you could include back trace from gdb? FreeRADIUS also includes tests for various authentication types, but it doesn't include tests with radclient for MS-CHAP -- You received this bug notification because you

[Bug 1081509] Re: freeradius + mysql fails to start

2022-02-21 Thread Alan DeKok
The solution is well documented in FreeRADIUS. Just change the "pool" section of the "sql" module configuration. Set "start = 0" -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1081509 Title:

[Bug 1955009] Re: Freeradius 3.0.21+dfsg-3build1 fails test of moonshot-gss-eap

2022-02-16 Thread Alan DeKok
We'll try to get it out this week. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1955009 Title: Freeradius 3.0.21+dfsg-3build1 fails test of moonshot-gss-eap To manage notifications about this bug

[Bug 1955009] Re: Freeradius 3.0.21+dfsg-3build1 fails test of moonshot-gss-eap

2022-01-07 Thread Alan DeKok
> So I now understand the OR change, just not why content_type is compared with UINT8_MAX. The TLS specification (RFC 8446, among others) says that the ContentType field is an 8-bit value. Therefore anything past that is not a real content type, and is "invented" by OpenSSL. -- You received

[Bug 1955009] Re: Freeradius 3.0.21+dfsg-3build1 fails test of moonshot-gss-eap

2022-01-07 Thread Alan DeKok
There are a LOT of changes required to get FreeRADIUS working with OpenSSL3. We will be releasing 3.0.26 in January to address these, and other issues. I'd suggest waiting for that. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu.

[Bug 1955009] Re: Freeradius 3.0.21+dfsg-3build1 fails test of moonshot-gss-eap

2021-12-17 Thread Alan DeKok
I would suggest trying 3.0.25. If that works, don't even bother trying to debug this. OpenSSL has minor behavior differences across a major version change. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu.

[Bug 1955009] Re: Freeradius 3.0.21+dfsg-3build1 fails test of moonshot-gss-eap

2021-12-16 Thread Alan DeKok
My $0.02 is to try the head of v3.0.x. I don't recall if we put in fixes specifically for OpenSSL 3, but perhaps. We've also *significantly* updated the TLS debugging output. It's a lot clearer, and gives a lot more information. -- You received this bug notification because you are a member

[Bug 1823053] Re: wpasupplicant 2.6 w/ openssl 1.1.1 triggers TLSv1.3 version intolerance on WPA2-Enterprise networks on Cosmic and Disco

2019-06-17 Thread Alan DeKok
> Sadly, there are RADIUS servers which suffer from TLS version intolerance and will refuse authentication when the client offers TLS 1.3 This statement is completely missing the point. There are *no standards available* for using TLS 1.3 with *any* EAP method. The IETF is working on them, but

[Bug 1795310] Re: FTBFS with MySQL 8.0

2018-10-01 Thread Alan DeKok
Fixes committed in v3.0.x: https://github.com/FreeRADIUS/freeradius- server/commit/2125490c92c29356cd1564080fe9dc1e05acf8e3 and in master branch: https://github.com/FreeRADIUS/freeradius- server/commit/1334cf5942dcf46fe0ad99a2381604b42a35b331 We won't do the v2.0.x branch, as that release is

[Bug 1730362] Re: fails to start as user/group freerad (recommended!)

2017-11-06 Thread Alan DeKok
> So it works, and I assume the freeradius documentation is not supposed to include explanations about a chroot? You're free to submit documentation. Either to the project itself, or to http://wiki.freeradius.org > Just for some reason, FreeBSD version works fine with user=freerad group=freerad

Re: [Bug 1613462] [NEW] Compatibility issues between freeradius 2.2.8 and openssl 1.0.2g

2016-08-16 Thread Alan DeKok
Ubuntu is free to create a patch themselves. For us, v2 is EOL, and has been EOL for over a year. We encourage everyone to upgrade to v3, which has all of the relevant fixes included. And, which has many more features. i.e. if you're going to upgrade OpenSSL to a new release, you might as

Re: [Bug 1294585] Re: Freeradius crashes on heavy load (with EAP)

2014-03-20 Thread Alan DeKok
You haven't said which version of the server this is for. If it's the ubuntu one, it's 2.1.12. That is YEARS out of date. There is no reason to track down a bug in a version that old. As the FreeRADIUS author, I suggest trying 2.2.4. It was released yesterday. -- You received this bug

Re: [Bug 1294585] Re: Freeradius crashes on heavy load (with EAP)

2014-03-20 Thread Alan DeKok
You haven't said which version of the server this is for. If it's the ubuntu one, it's 2.1.12. That is YEARS out of date. There is no reason to track down a bug in a version that old. As the FreeRADIUS author, I suggest trying 2.2.4. It was released yesterday. -- You received this bug

Re: [Bug 1090365] [NEW] fails to build from source

2012-12-14 Thread Alan DeKok
Maxxer wrote: libtool: compile: 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

Re: [Bug 1090365] [NEW] fails to build from source

2012-12-14 Thread Alan DeKok
Maxxer wrote: libtool: compile: 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

Re: [Bug 938044] [NEW] Unable to listen on IPv6

2012-02-21 Thread Alan DeKok
Mark Schouten wrote: It is impossible to bind freeradius in Ubuntu to :: (ipv6 any). This seems to be a known bug for about three years [1]. Please look into this. Might be nice if this is fixed before Precise, which will definitly run in IPv6 only environments in it's support lifecycle. It

Re: [Bug 938044] [NEW] Unable to listen on IPv6

2012-02-21 Thread Alan DeKok
Mark Schouten wrote: It is impossible to bind freeradius in Ubuntu to :: (ipv6 any). This seems to be a known bug for about three years [1]. Please look into this. Might be nice if this is fixed before Precise, which will definitly run in IPv6 only environments in it's support lifecycle. It

Re: [Bug 283807] Re: freeradius with openssl support doesn't compile

2010-05-07 Thread Alan DeKok
Stas Sușcov wrote: Anybody knows if it's worth rebuilding the package using lucid sources? Otherwise I'm leaving it like it is. Later versions of FreeRADIUS have fixes. I suggest leaving the current 2.1.0 version alone, and ugprading instead. -- freeradius with openssl support doesn't

Re: [Bug 283807] Re: freeradius with openssl support doesn't compile

2010-05-07 Thread Alan DeKok
Stas Sușcov wrote: Anybody knows if it's worth rebuilding the package using lucid sources? Otherwise I'm leaving it like it is. Later versions of FreeRADIUS have fixes. I suggest leaving the current 2.1.0 version alone, and ugprading instead. -- freeradius with openssl support doesn't

[Bug 430732] Re: radclient doesn't work

2009-09-17 Thread Alan DeKok
Ah... localhost likely means ::1, or IPv6 transport. If the OS supplies the header files, but doesn't have the code to implement the feature, then that's likely the error you'll get. The simplest thing to do is to disable udpfromto via the configure script. -- radclient doesn't work

[Bug 430732] Re: radclient doesn't work

2009-09-17 Thread Alan DeKok
Ah... localhost likely means ::1, or IPv6 transport. If the OS supplies the header files, but doesn't have the code to implement the feature, then that's likely the error you'll get. The simplest thing to do is to disable udpfromto via the configure script. -- radclient doesn't work

[Bug 421005] Re: can't compile version from repository (on 64bit)

2009-09-05 Thread Alan DeKok
It's a bug in libltdl that is in Ubuntu. See /usr/include/ltdl.h on the affected system. It has a line: #define lt_preloaded_symbolslt__PROGRAM__LTX_preloaded_symbols Q: Where is lt__PROGRAM__LTX_preloaded_symbols defined or referenced? A: Nowhere. If you spend some time spelunking

[Bug 421005] Re: can't compile version from repository (on 64bit)

2009-09-05 Thread Alan DeKok
It's a bug in libltdl that is in Ubuntu. See /usr/include/ltdl.h on the affected system. It has a line: #define lt_preloaded_symbolslt__PROGRAM__LTX_preloaded_symbols Q: Where is lt__PROGRAM__LTX_preloaded_symbols defined or referenced? A: Nowhere. If you spend some time spelunking

[Bug 421005] Re: can't compile version from repository (on 64bit)

2009-09-04 Thread Alan DeKok
It looks like a conflict between the ltdl.h / libltld.a included on the system, and the ones included with the server. Due to magic issues with libtool, it can compile using the local files, and then link to the global files. Hence the conflict. In version 2.1.7 (to be released soon), we've

[Bug 421005] Re: can't compile version from repository (on 64bit)

2009-09-04 Thread Alan DeKok
It looks like a conflict between the ltdl.h / libltld.a included on the system, and the ones included with the server. Due to magic issues with libtool, it can compile using the local files, and then link to the global files. Hence the conflict. In version 2.1.7 (to be released soon), we've

Re: [Bug 367769] Re: freeradius init script not consistent with radiusd.conf

2009-04-27 Thread Alan DeKok
Thierry Carrez wrote: ** Changed in: freeradius (Ubuntu) Importance: Undecided = High ** Changed in: freeradius (Ubuntu) Status: New = Confirmed ** Changed in: freeradius (Ubuntu) Assignee: (unassigned) = Thierry Carrez (ttx) See also later versions of FreeRADIUS. This

Re: [Bug 367769] Re: freeradius init script not consistent with radiusd.conf

2009-04-27 Thread Alan DeKok
Thierry Carrez wrote: ** Changed in: freeradius (Ubuntu) Importance: Undecided = High ** Changed in: freeradius (Ubuntu) Status: New = Confirmed ** Changed in: freeradius (Ubuntu) Assignee: (unassigned) = Thierry Carrez (ttx) See also later versions of FreeRADIUS. This

[Bug 315896] Re: freeradius upgrade broken in hardy backports

2009-01-12 Thread Alan DeKok
OK. I've put similar patches into git head (stable master branches). They should be in the next major release of the server (2.1.4) -- freeradius upgrade broken in hardy backports https://bugs.launchpad.net/bugs/315896 You received this bug notification because you are a member of Ubuntu Bugs,

[Bug 315896] Re: freeradius upgrade broken in hardy backports

2009-01-10 Thread Alan DeKok
Performing automated upgrades across major revision numbers of a program isn't recommended. There are a large number of changes in the configuration files and in their processing from version 1.1.x to version 2.1.x. Writing scripts to do automated upgrades is extremely difficult. The upgrade

[Bug 315896] Re: freeradius upgrade broken in hardy backports

2009-01-10 Thread Alan DeKok
Performing automated upgrades across major revision numbers of a program isn't recommended. There are a large number of changes in the configuration files and in their processing from version 1.1.x to version 2.1.x. Writing scripts to do automated upgrades is extremely difficult. The upgrade

Re: [Bug 120024] Re: libperl.so doesn't exist, and perl -MExtUtils::Embed -e ldopts outputs -lperl

2007-09-19 Thread Alan DeKok
not only at the problem, but at the attitude of the package maintainers who seem to be saying just step through a bunch of poorly documented hoops to work around our lying package. Alan DeKok. -- libperl.so doesn't exist, and perl -MExtUtils::Embed -e ldopts outputs -lperl https

[Bug 120024] libperl.so doesn't exist, and perl -MExtUtils::Embed -e ldopts outputs -lperl

2007-06-12 Thread Alan DeKok
Public bug reported: Binary package hint: libperl5.8 $ locate libperl ... /var/lib/dpkg/info/libperl5.8.shlibs /usr/lib/libperl.so.5.8 /usr/lib/libperl.so.5.8.8 /usr/share/defoma/libperl-file.pl ... There's no libperl.a, and no libperl.so $ perl -MExtUtils::Embed -e ldopts -Wl,-E