[
https://issues.apache.org/jira/browse/AXIS2C-1342?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Robert Lazarski resolved AXIS2C-1342.
-------------------------------------
Fix Version/s: 2.0.0
(was: 1.7.0)
Resolution: Fixed
Two issues caused crashes when persistOperationContext was enabled
under high concurrency:
1. Thread-safety issue in axis2_conf_ctx_fill_ctxs: The function
directly accessed conf_ctx->svc_grp_ctx_map without mutex locking,
while other functions used the mutex. This caused race conditions
when multiple threads accessed the map concurrently. Fixed by using
the thread-safe axis2_conf_ctx_get_svc_grp_ctx accessor.
2. Allocator pool leak in ctx_handler: When axis2_op_ctx_create failed
after switching to global pool, the function returned without
switching back to local pool. This left subsequent allocations
happening in the wrong pool. Fixed by switching back to local pool
before returning on failure.
> When enable "persistOperationContext" parameter in axis2.xml, server crashes.
> ------------------------------------------------------------------------------
>
> Key: AXIS2C-1342
> URL: https://issues.apache.org/jira/browse/AXIS2C-1342
> Project: Axis2-C
> Issue Type: Bug
> Components: core/description, core/engine
> Environment: WinXP, possibly Linux as well
> Reporter: Selvaratnam Uthaiyashankar
> Priority: Major
> Fix For: 2.0.0
>
>
> When enable <parameter name="persistOperationContext"
> locked="false">true</parameter> in axis2.xml and send large number of
> requests (10000 in 10 concurrent requests) using ab, httpd crashes. (It
> seems, simple axis2 server does NOT crashes)
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
