Okay,
I did:
rm /usr/local/lib/*
cd radiusd (root dir of src of 0.4)
make distclean
./configure --sysconfdir=/etc --localstatedir=/var --with-threads \
--with-thread-pool \
--with-rlm-sql-postgresql-include-dir=/opt/pgsql/include \
--with-rlm-sql-postgresql-lib-dir=/opt/pgsql/lib
make
make install
gdb radiusd
(gdb) set args -X
(gdb) run
(ran a few requests)
Seg. Faulted
(gdb)bt
This is gdb out of 'bt'
+++ START
#0 0x4001a96d in sql_userparse () at eval.c:88
#1 0x4001aa61 in sql_getvpdata () at eval.c:88
#2 0x40019ae2 in rlm_sql_authorize () at eval.c:88
#3 0x80550a6 in call_modsingle (component=1, sp=0x80d5eb0,
request=0x80dcdd8,
default_result=6) at modcall.c:205
#4 0x80551fe in modcall (component=1, c=0x80d5eb0, request=0x80dcdd8)
at modcall.c:288
#5 0x80550f5 in call_modgroup (component=1, g=0x80d5e00,
request=0x80dcdd8,
default_result=6) at modcall.c:227
#6 0x80551bf in modcall (component=1, c=0x80d5e00, request=0x80dcdd8)
at modcall.c:281
#7 0x80549a6 in indexed_modcall (comp=1, idx=0, request=0x80dcdd8)
at modules.c:456
#8 0x8054cd5 in module_authorize (request=0x80dcdd8) at modules.c:631
#9 0x8051b4c in rad_authenticate (request=0x80dcdd8) at auth.c:524
#10 0x804d763 in rad_respond (request=0x80dcdd8,
fun=0x80519f0 <rad_authenticate>) at radiusd.c:1492
#11 0x804d3c3 in rad_process (request=0x80dcdd8, dospawn=0) at
radiusd.c:1252
#12 0x804d09b in main (argc=2, argv=0xbffff994) at radiusd.c:1060
#13 0x400ae2eb in __libc_start_main (main=0x804c3ec <main>, argc=2,
ubp_av=0xbffff994, init=0x804b4b8 <_init>, fini=0x805deac <_fini>,
rtld_fini=0x4000c130 <_dl_fini>, stack_end=0xbffff98c)
at ../sysdeps/generic/libc-start.c:129
+++++END
On Fri, 14 Dec 2001, Chris Parker wrote:
> At 02:03 PM 12/14/2001 -0800, [EMAIL PROTECTED] wrote:
>
> >I've been having a seg fault in snapshots and in 0.4, I don't get it in
> >0.3. Basically I run about 5 request and then it seg faults. Not exactly
> >5, but aprox. The only auth I do is through SQL and it then runs a simple
> >external script that returns a Reply-Message and then exits 0.
> >Running 'radiusd -xx' doesn't seem to give any info on why.
> >Where do I start to get this debugged? What additional info do we need?
>
> First thing to suggest, is remove the 'rlm_*' files in /usr/local/lib.
>
> Most likely you have an older lib being linked in, and some of the
> function calls to that lib have changed.
>
> If you do this, and the server still segfaults, run it via the
> following:
>
> system$ gdb radiusd
> ...
> (gdb) set args -X
> (gdb) run
>
> When/If the server segfaults type:
>
> (gdb) bt
>
> And send the results of the server debug info and the gdb stack trace
> to this list.
>
> Thanks,
> -Chris
> --
> \\\|||/// \ Chris Parker - Manager, Development Engineering
> \ ~ ~ / \ WX *is* Wireless! \ [EMAIL PROTECTED]
> | @ @ | \ http://www.starnetwx.net \ (847) 963-0116
> oOo---(_)---oOo--\------------------------------------------------------
> \ Without C we would have 'obol', 'basi', and 'pasal'
>
>
> -
> List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
>
========================================================================
Check out http://www.setfree.org It can change your life.
========================================================================
-
List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
