[ 
https://issues.apache.org/jira/browse/AXIS2C-1192?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Robert Lazarski resolved AXIS2C-1192.
-------------------------------------
    Fix Version/s: 2.0.0
                       (was: 1.7.0)
       Resolution: Fixed

  Root Cause: Two memory leaks in listener_manager.c caused heap corruption 
over repeated runs:

  1. arg_list leak (src/core/clientapi/listener_manager.c:134): The 
axis2_listener_manager_worker_func_args_t structure was allocated when starting 
the listener thread but never freed.
  2. tl_state leak (src/core/clientapi/listener_manager.c:231): When 
axis2_listener_manager_stop() was called and waiting_calls dropped to 0, the 
tl_state pointer was set to NULL without freeing the memory.

  Fix (Commit 83bfc7c1c):
  - listener_manager.c:298-301: Save listener pointer, then free arg_list in 
worker function
  - listener_manager.c:232-234: Free tl_state before setting pointer to NULL in 
stop function

  These memory leaks accumulated on each client run, eventually corrupting the 
heap and causing crashes on subsequent executions.


> axis2c dual sample clients give a memory corruption in consecutive execution
> ----------------------------------------------------------------------------
>
>                 Key: AXIS2C-1192
>                 URL: https://issues.apache.org/jira/browse/AXIS2C-1192
>             Project: Axis2-C
>          Issue Type: Bug
>          Components: samples
>         Environment: Windows XP, guththila
>            Reporter: Selvaratnam Uthaiyashankar
>            Priority: Major
>             Fix For: 2.0.0
>
>
> axis2c dual sample clients give a memory corruption in consecutive execution.
> To recreate the issue, try running echo_blocking_dual.exe several times 
> consecutively. The corruption occurs randomly.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to