Vamsi Pavan Kumar Sanka just shared SANDESHA2C-75 with you
---------------------------------------------------
> Crash in Sandesha2/C with Sqlite persistence after 174 simple messages
> consistently.
> ------------------------------------------------------------------------------------
>
> Key: SANDESHA2C-75
> URL: https://issues.apache.org/jira/browse/SANDESHA2C-75
> Project: Sandesha2/C
> Issue Type: Bug
> Environment: Linux64/ Axis2C (1.6.0) + Sandesha2C (0.91)
> Reporter: Vamsi Pavan Kumar Sanka
> Labels: crash, sandesha2, sqlite
>
> While testing the standalone sample program for sandesha2/c for sending 200
> messages (basic echo message), we could see crash consistently at following
> location after lot of debugging.
> file - src/storage/sqlite/permanent_seq_property_mgr.c +344
> axis2_bool_t AXIS2_CALL
> sandesha2_permanent_seq_property_mgr_update(
> sandesha2_seq_property_mgr_t *seq_prop_mgr,
> const axutil_env_t *env,
> sandesha2_seq_property_bean_t *bean)
> {
> axis2_char_t sql_update[1024];
> axis2_bool_t ret = AXIS2_FALSE;
> axis2_char_t *id = NULL;
> sandesha2_permanent_seq_property_mgr_t *seq_prop_mgr_impl = NULL;
> axis2_char_t *seq_id = NULL;
> axis2_char_t *name = NULL;
> axis2_char_t *value = NULL;
> AXIS2_PARAM_CHECK(env->error, bean, AXIS2_FALSE);
> id = sandesha2_permanent_seq_property_mgr_get_id_with_bean(env, bean);
> seq_id = sandesha2_seq_property_bean_get_seq_id(bean, env);
> name = sandesha2_seq_property_bean_get_name(bean, env);
> value = sandesha2_seq_property_bean_get_value(bean, env);
> seq_prop_mgr_impl = SANDESHA2_INTF_TO_IMPL(seq_prop_mgr);
> sprintf(sql_update, "update seq_property set seq_id='%s', name='%s',"\
> "value='%s' where id='%s'", seq_id, name, value, id);
> if(id)
> AXIS2_FREE(env->allocator, id);
> ret = sandesha2_permanent_bean_mgr_update(seq_prop_mgr_impl->bean_mgr,
> env,
> sql_update);
> return ret;
> }
> Crash seen at sprintf line.
> sql_update variable size is fixed as 1024. After certain number of messages
> (in our case, 176 messages exactly), sprintf crashes with buffer overflow as
> value is keep appending the messages number (it grows as we continue further).
> I hope I made everything clear. Now, i want to know whether this is known
> issue which got fixed or if not, can anybody fix this ?
> Unfortunately, for 0.91 version (which is downloadable from apache website)
> there is no provision for inmemory storage mgr (to use it as alternative
> case), as code is hard coded to persistence based storage mgr.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
