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

Re: (ITS#4117) slapd-ldap killed by a broken client



One preliminary suggestion is to try 2.3.11, because lot of work
occurred on libldap concurrency issues which is very unlikely to be
backported to 2.2.  Note that 2.3 proxying (slapd-ldap/meta, slapo-rwm
and so) received a lot of attention recently, and it's far better than
2.2 ever.  If you can't reproduce the issue with 2.3, then you should
definitely consider updating, at least the proxy.

p.

On Wed, 2005-11-02 at 09:07 +0000, raphael.ouazana@linagora.com wrote:
> On Sam 29 octobre 2005 10:46, raphael.ouazana@linagora.com wrote:
> > Ok, I think SIGPIPE appears because of gdb. So I'll try to reproduce the
> > bug with a correct signal handling of gdb.
> 
> This seems to be the correct trace. It occurs after the SIGPIPE is ignored :
> 
> Program received signal SIGSEGV, Segmentation fault.
> 
> (gdb) bt
> #0  0x080cd835 in ldap_int_select (ld=0x83079c8, timeout=0xb7090690)
>     at os-ip.c:747
> #1  0x080c014f in wait4msg (ld=0x83079c8, msgid=34, all=0,
> timeout=0xb7090768,
>     result=0xb7090734) at result.c:331
> #2  0x080bfdeb in ldap_result (ld=0x83079c8, msgid=34, all=0,
>     timeout=0xb7090768, result=0xb7090734) at result.c:126
> #3  0x0809ccb5 in ldap_back_search (op=0x83088e0, rs=0xb7091870)
>     at search.c:181
> #4  0x08058b6b in do_search (op=0x83088e0, rs=0xb7091870) at search.c:412
> #5  0x0805745a in connection_operation (ctx=0xb70918f0, arg_v=0x83088e0)
>     at connection.c:1079
> #6  0x080befab in ldap_int_thread_pool_wrapper (xpool=0x82b9d00) at
> tpool.c:467
> #7  0x00bfede8 in start_thread () from /lib/tls/libpthread.so.0
> #8  0x00f0b93a in clone () from /lib/tls/libc.so.6
> 
> Raphael Ouazana.
> 
> 



    SysNet - via Dossi,8 27100 Pavia Tel: +390382573859 Fax: +390382476497