I had exactly the same problem with SER 0.9.6 and radiusclient-ng
0.5.2 on FreeBSD/Sparc64. I read all the problem reports I found on
Google - they all led to the suggestion that my shared secrets
between radiusclient and FreeRadius mismatch. However, I double-
checked the entries - I also set them to some very simple values, all
to no avail.
So I cloned the Sparc64 setup to an i386 machine using the exact same
releases of FreeBSD, SER and radiusclient-ng. I scp'ed all the config
files from the Spacr box to the i386...
and presto, it works like a charm. So maybe it really is a problem
with calculation of MD5 on 64bit architectures.
cheers
Heimo
(sorry, I don't know how to debug the source myself, but if I can be
of any help leading to a fix, please let me know!)
Am 09.03.2006 um 23:36 schrieb Jan Janak:
Do you have the same shared secret configured in the
client library
and server ?
Jan.
Edson wrote:
I'm facing some really weird problem.
Let's try to explain, but
first my
config (basically):
OpenSUSE 10.0
FreeRadius 1.0.4-4
RadiusClient-NG 0.5.2
SER 0.9.6
MySQL 5.0.18
THE GOALs: upgrade from SER 0.8.4 to 0.9.6; MySQL from 4.0 to 5.0.
THE SCENE: FreeRadius, using MySQL as back-end, as SER. SER
configured to
consult Radius and make account on both places (Radius and MySQL).
I have
this same configuration running and OK, but on an old version of
SER (0.8.4)
+ MySQL 4.0 + RadiusClient-NG 0.5.0. MySQL is running OK and
responding to
all queries, as expected.
THE PROBLEM: with the upgrade the RadiusClient-NG is reporting
that the
digest (MD5) returned by the Radius server isn't correct. If I
tweak the
code of sendserver.c (radiusclient-ng-0.5.2/lib/sendserver.c) to
compile it
with DEBUG displays, it shows me that different digests. One that
comes from
FreeRadius and another calculated.
So now I stuck... I can not go on with the upgrade 'til I find a
solution to
this issue. Not using Radius is not a possibility.
Did anybody cross this problem and find a solution?
Edson.
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers
_______________________________________________
Serusers mailing list
serusers(a)lists.iptel.org
http://lists.iptel.org/mailman/listinfo/serusers