[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: (ITS#5235) ppolicy leads to segfault



ando@sys-net.it wrote:
> dieter@dkluenter.de wrote:
> 
>> (gdb) bt
>> #0  0xb7a67139 in free () from /lib/libc.so.6
>> #1  0xb7ec128a in ber_memfree_x (p=0x2001, ctx=0x0) at memory.c:152
>> #2  0x0808f556 in ch_free ()
>> #3  0xb76f458c in ppolicy_restrict (op=0xa0a50d54, rs=0xa0a51148)
>>     at ppolicy.c:1245
> 
> I suspect this free occurs an internal operation initiated with
> connection_fake_init(); can you check, from the core, if the value of
> op->o_conn->c_conn_idx is -1?

If that's the real problem, the better fix would be just to make ppolicy 
ignore fake connections. Certainly internal operations can never be restricted.
-- 
   -- Howard Chu
   Chief Architect, Symas Corp.  http://www.symas.com
   Director, Highland Sun        http://highlandsun.com/hyc/
   Chief Architect, OpenLDAP     http://www.openldap.org/project/