I am OK with patch except for the change mentioned below. But
avnd_evt_clc_resp_evh() can be simplified and made more elegant now when
cmd_type is available. I will send an additional patch that should be
merged with this one.
/Hans
On 08/07/2013 05:03 PM, Hans Feldt wrote:
> On 7 August 2013 14:14, <praveen.malv...@oracle.com> wrote:
>> osaf/services/saf/avsv/avnd/avnd_clc.c | 93
>> +++++++++++--------------
>> osaf/services/saf/avsv/avnd/include/avnd_evt.h | 2 -
>> 2 files changed, 40 insertions(+), 55 deletions(-)
>>
>>
>> At present when clc response event comes AMFND searches whole component
>> database and matches the context. This context does not include component
>> name. Thus AMFND can associate clc response with wrong component. The patch
>> ensures that AMFND gets component name in the clc response event and thus
>> solves the problem by associating clc response event with the component.
>>
>> diff --git a/osaf/services/saf/avsv/avnd/avnd_clc.c
>> b/osaf/services/saf/avsv/avnd/avnd_clc.c
>> --- a/osaf/services/saf/avsv/avnd/avnd_clc.c
>> +++ b/osaf/services/saf/avsv/avnd/avnd_clc.c
>> @@ -70,7 +70,6 @@ static uint32_t avnd_comp_clc_orph_resta
>>
>> uint32_t avnd_comp_clc_st_chng_prc(AVND_CB *, AVND_COMP *,
>> SaAmfPresenceStateT, SaAmfPresenceStateT);
>>
>> -static uint32_t avnd_comp_clc_resp(NCS_OS_PROC_EXECUTE_TIMED_CB_INFO *);
>> static uint32_t avnd_instfail_su_failover(AVND_CB *, AVND_SU *, AVND_COMP
>> *);
>>
>>
>> /***************************************************************************
>> @@ -311,33 +310,20 @@ uint32_t avnd_evt_clc_resp_evh(AVND_CB *
>> AVND_CLC_EVT *clc_evt = &evt->info.clc;
>> AVND_COMP *comp = 0;
>> uint32_t rc = NCSCC_RC_SUCCESS;
>> - uint32_t amcmd = 0;
>> TRACE_ENTER();
>>
>> - /* get the comp */
>> - if (clc_evt->exec_ctxt) {
>> - /* => cmd successfully launched (scan the entire
>> comp-database) */
>> - for (comp = (AVND_COMP
>> *)ncs_patricia_tree_getnext(&cb->compdb,
>> - (uint8_t
>> *)0); comp;
>> - comp = (AVND_COMP
>> *)ncs_patricia_tree_getnext(&cb->compdb, (uint8_t *)&comp->name)) {
>> - if (comp->clc_info.cmd_exec_ctxt ==
>> clc_evt->exec_ctxt)
>> - break;
>> - if (comp->clc_info.am_cmd_exec_ctxt ==
>> clc_evt->exec_ctxt) {
>> - amcmd = 1;
>> - break;
>> - }
>> - }
>> - } else
>> - /* => cmd did not launch successfully (comp is available) */
>> - comp = m_AVND_COMPDB_REC_GET(cb->compdb, clc_evt->comp_name);
>> + comp = m_AVND_COMPDB_REC_GET(cb->compdb, clc_evt->comp_name);
>>
>> /* either other cmd is currently executing or the comp is deleted */
>> - if (!comp)
>> - goto done;
>> + if (comp == NULL) {
>> + LOG_WA("%s: could not find comp '%s'", __FUNCTION__,
>> + clc_evt->comp_name.value);
>> + goto done;
>> + }
>>
>> TRACE("'%s'", comp->name.value);
>>
>> - if (amcmd || clc_evt->cmd_type == AVND_COMP_CLC_CMD_TYPE_AMSTART ||
>> + if (clc_evt->cmd_type == AVND_COMP_CLC_CMD_TYPE_AMSTART ||
>> clc_evt->cmd_type == AVND_COMP_CLC_CMD_TYPE_AMSTOP) {
>> TRACE("CLC command
>> type:'%s'",clc_cmd_type[clc_evt->cmd_type]);
>> switch (comp->clc_info.am_exec_cmd) {
>> @@ -431,7 +417,7 @@ uint32_t avnd_evt_clc_resp_evh(AVND_CB *
>> break;
>>
>> default:
>> - osafassert(0);
>> + goto done;
>
> Why this change? It will silently discard the event.
> This is a typical case of assert (not osafassert) isn't it?
> /Hans
>
>
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite!
It's a free troubleshooting tool designed for production.
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
_______________________________________________
Opensaf-devel mailing list
Opensaf-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/opensaf-devel
