Hi, Can you run using simple axis server and see whether it crashes? When you are debugging httpd with -X flag, it runs in a single process. That might be the reason why it works.. If it crash it simple axis server, it might be easy to debug.
Regards, Shankar On Wed, Sep 23, 2009 at 7:53 PM, Jose M Medina V <jose_med...@wesamconsulting.com> wrote: > Any ideas of how fix this problem? > Regards, > Begin forwarded message: > > From: Jose M Medina V <jose_med...@wesamconsulting.com> > Date: September 21, 2009 8:47:51 PM GMT-04:30 > To: "Apache AXIS C User List" <axis-c-user@ws.apache.org> > Subject: Re: Problem with Ubuntu > > Well here an update of the problem > I could reproduce the error in a Centos 5 (64 bits) installation, so I think > the problem is something related with 64 bit installation and not with the > deferent operating system. > As you suggest, I install the valgrind utilities and I run the test with the > following command > [r...@linux-desa logs]# valgrind /usr/sbin/httpd -X -d /etc/httpd > ==23257== Memcheck, a memory error detector > ==23257== Copyright (C) 2002-2009, and GNU GPL'd, by Julian Seward et al. > ==23257== Using Valgrind-3.5.0 and LibVEX; rerun with -h for copyright info > ==23257== Command: /usr/sbin/httpd -X -d /etc/httpd > ==23257== > ==23257== Syscall param semctl(IPC_SET, arg.buf) points to uninitialised > byte(s) > ==23257== at 0x69AA4CA: semctl (in /lib64/libc-2.5.so) > ==23257== by 0x144E70: unixd_set_proc_mutex_perms (in /usr/sbin/httpd) > ==23257== by 0x144E98: unixd_set_global_mutex_perms (in /usr/sbin/httpd) > ==23257== by 0xE2AB6FA: ??? (in /usr/lib64/httpd/modules/mod_rewrite.so) > ==23257== by 0x13186B: ap_run_post_config (in /usr/sbin/httpd) > ==23257== by 0x11ECD2: main (in /usr/sbin/httpd) > ==23257== Address 0x7ff000640 is on thread 1's stack > ==23257== > ==23257== Syscall param semctl(IPC_SET, arg.buf) points to uninitialised > byte(s) > ==23257== at 0x69AA4CA: semctl (in /lib64/libc-2.5.so) > ==23257== by 0x144E70: unixd_set_proc_mutex_perms (in /usr/sbin/httpd) > ==23257== by 0x144E98: unixd_set_global_mutex_perms (in /usr/sbin/httpd) > ==23257== by 0xAE06B0D: ??? (in /usr/lib64/httpd/modules/mod_ldap.so) > ==23257== by 0x13186B: ap_run_post_config (in /usr/sbin/httpd) > ==23257== by 0x11EE1E: main (in /usr/sbin/httpd) > ==23257== Address 0x7ff000640 is on thread 1's stack > ==23257== > ==23257== Syscall param semctl(IPC_SET, arg.buf) points to uninitialised > byte(s) > ==23257== at 0x69AA4CA: semctl (in /lib64/libc-2.5.so) > ==23257== by 0x144E70: unixd_set_proc_mutex_perms (in /usr/sbin/httpd) > ==23257== by 0x1440D2: ap_mpm_run (in /usr/sbin/httpd) > ==23257== by 0x11EE47: main (in /usr/sbin/httpd) > ==23257== Address 0x7ff000570 is on thread 1's stack > The result was the web service work perfect, but when I start the apache > server as normal, I still get the core in the apache log file. > Regards, > > On Sep 21, 2009, at 2:44 PM, Murphey McCloy wrote: > > Just place valgrind in front of your httpd command (adding the -X parameter) > and run it. For example, in my configuration I do: > valgrind /usr/sbin/httpd -X -d /etc/httpd > Then let it run and try to make it seg fault. The output it displays should > be useful. > > > On Sep 21, 2009, at 11:28 AM, "Jose M Medina V" > <jose_med...@wesamconsulting.com> wrote: > > I didn’t try with this tool because I don’t know until now. > I will read the documentation of Valgrind and run some test in the Ubuntu > configuration. In the meantime, anybody has some examples for use this > tool?? Any help would be appreciated. > Regards > On Sep 21, 2009, at 11:48 AM, Murphey McCloy wrote: > > Hey there. Have you tried running your application via valgrind to see if > there is memory corruption or access violations? I realize that it is > working on CentOS but not on Ubuntu. The fact that it works on CentOS > without crashing doesn’t mean there aren’t memory issues, it very well may > be that you have some uninitialized value being referenced that is causing > the seg fault. CentOS might be setting the memory to 0 by default, while > Ubuntu might not. > > > From: Jose M Medina V [mailto:jose_med...@wesamconsulting.com] > Sent: Monday, September 21, 2009 10:03 AM > To: Apache AXIS C User List > Subject: Fwd: Problem with Ubuntu > > Any update with this issue? > > Thanks, > > Begin forwarded message: > > From: Jose M Medina V <jose_med...@wesamconsulting.com> > Date: September 17, 2009 8:26:24 AM GMT-04:30 > To: "Apache AXIS C User List" <axis-c-user@ws.apache.org> > Subject: Re: Problem with Ubuntu > > Selvaratnam thank for your response. > > I’m attaching all source code of the web service; the function > axis2_srvprb_agrApuesta is in web_serv_prb.c library. > > As additional information the command that I use to build axis2c in Ubuntu > was: > ./configure --prefix=/opt/axis2c --with-apache2=/usr/include/apache2 \ > --enable-libxml2=yes --with-apr=/usr/include/apr-1.0 --enable-openssl=yes > \ > --enable-trace=yes > > Also the command that I use in Centos 5-32 bit (ambient where the test run > perfect) was: > ./configure --prefix=/opt/axis2c --with-apache2=/usr/include/httpd > --with-apr=/usr/include/apr-1\ > --enable-libxml2=yes --enable-openssl=yes \ > --enable-trace=yes > > Again thank in advance. > > On Sep 16, 2009, at 10:31 PM, Selvaratnam Uthaiyashankar wrote: > > Hi, > > Can you give "axis2_srvprb_agrApuesta" method as well? I think the > problem might be in the return value, because srvprb_invoke exited > without any problem. > > Regards, > Shankar > > > On Wed, Sep 16, 2009 at 8:59 PM, Jose M Medina V > <jose_med...@wesamconsulting.com> wrote: > > I install Axis2c into Ubuntu Server 9.04 (Linux ubuntu-server > > 2.6.28-15-server #49-Ubuntu SMP Tue Aug 18 20:09:37 UTC 2009 x86_64 > > GNU/Linux) as a module for Apache 2. After that I write a web service for > > test purpose and when I test the service I receive the following error in > > the Apache log. > > “[Tue Sep 15 19:57:45 2009] [notice] child pid 3371 exit signal Segmentation > > fault (11)” > > I start debug the program and I determine that the error was on the return > > of the funtion srvprb_invoke, see the following axis2.log trace. > > Tue Sep 15 19:57:44 2009] [debug] web_serv_prb.c(323) [pruebaSrv] > > axis2_srvprb_agrApuesta:: Fin de Funcion > > [Tue Sep 15 19:57:44 2009] [debug] lib_skeleton.c(118) [pruebaSrv] > > srvprb_invoke:: Fin de Funcion > > The complete funtion srvprb_invoke is: > > axiom_node_t *AXIS2_CALL > > srvprb_invoke( > > axis2_svc_skeleton_t * svc_skeleton, > > const axutil_env_t * env, > > axiom_node_t * node, > > axis2_msg_ctx_t * msg_ctx) > > { > > char *nombreFuncion="srvprb_invoke::"; > > axiom_node_t *node_srvprb_agrApuesta=NULL; > > > > AXIS2_LOG_DEBUG(env->log, AXIS2_LOG_SI, "[pruebaSrv] %s Inicio de > > Funcion",nombreFuncion); > > /* Dependiendo de la operacion que llegue se llama a la respectiva > > funcion que procesa el > > * request > > */ > > if (node) > > { > > if (axiom_node_get_node_type(node, env) == AXIOM_ELEMENT) > > { > > axiom_element_t *element = NULL; > > element = > > (axiom_element_t *) axiom_node_get_data_element(node, env); > > if (element) > > { > > axis2_char_t *op_name = > > axiom_element_get_localname(element, env); > > if (op_name) > > { > > if (axutil_strcmp(op_name, "inserta_apuesta") == 0) > > { > > node_srvprb_agrApuesta = axis2_srvprb_agrApuesta(env, > > node); > > if(node_srvprb_agrApuesta != NULL) > > { > > AXIS2_LOG_DEBUG(env->log, AXIS2_LOG_SI, "[pruebaSrv] > > %s Fin de Funcion",nombreFuncion); > > return node_srvprb_agrApuesta; > > } > > else > > { > > AXIS2_LOG_ERROR(env->log, AXIS2_LOG_SI, "[pruebaSrv] > > %s Resultado en NULL",nombreFuncion); > > return NULL; > > } > > } > > } > > } > > } > > } > > AXIS2_LOG_ERROR(env->log, AXIS2_LOG_SI, "[pruebaSrv] %s Parametro OM > > invalido en el request",nombreFuncion); > > AXIS2_ERROR_SET(env->error, > > AXIS2_ERROR_SVC_SKEL_INVALID_XML_FORMAT_IN_REQUEST, > > AXIS2_FAILURE); > > return NULL; > > } > > Also I run the same test into a Centos 5 installation and the test was > > perfect, so I think the problem is something about ubuntu. > > Could anyone help me with this problem? What can I revise? > > Thanks in advance > > ------------------------------------------------------------ > > José M. Medina V. > > Wesam Consulting > > jose_med...@wesamconsulting.com > > Mobile: +58(416)613-3302 +58(414)019-0580 > > Ofc: +58(212)959-4155 Ext. 204 > > BB pin: 31E562BF > > Caracas - Venezuela > > > > > > > > > > > > > > > -- > S.Uthaiyashankar > Software Architect > WSO2 Inc. > http://wso2.com/ - "The Open Source SOA Company" > > ------------------------------------------------------------ > José M. Medina V. > Wesam Consulting > jose_med...@wesamconsulting.com > Mobile: +58(416)613-3302 +58(414)019-0580 > Ofc: +58(212)959-4155 Ext. 204 > BB pin: 31E562BF > Caracas - Venezuela > > > ------------------------------------------------------------ > José M. Medina V. > Wesam Consulting > jose_med...@wesamconsulting.com > Mobile: +58(416)613-3302 +58(414)019-0580 > Ofc: +58(212)959-4155 Ext. 204 > BB pin: 31E562BF > Caracas - Venezuela > > > > > > > ------------------------------------------------------------ > José M. Medina V. > Wesam Consulting > jose_med...@wesamconsulting.com > Mobile: +58(416)613-3302 +58(414)019-0580 > Ofc: +58(212)959-4155 Ext. 204 > BB pin: 31E562BF > Caracas - Venezuela > > > > > > > ------------------------------------------------------------ > José M. Medina V. > Wesam Consulting > jose_med...@wesamconsulting.com > Mobile: +58(416)613-3302 +58(414)019-0580 > Ofc: +58(212)959-4155 Ext. 204 > BB pin: 31E562BF > Caracas - Venezuela > > > > > > -- S.Uthaiyashankar Software Architect WSO2 Inc. http://wso2.com/ - "The Open Source SOA Company"
