I just switched to multi-threading, and I get an Assertion on shutdown:
Here is the trace:
nua: nua_shutdown: entering
nua((nil)): signal r_shutdown
nua: nua_stack_shutdown: entering
Time 14:33:27: sipuad: nua_shutdown() started
tport_destroy(0x10268418)
nua: nua_stack_event: entering
nua((nil)):
Thanks for the replies.
I tried putting more su_root_step() calls in my application, and this
makes things much better, but I am still seeing timer E expiring even
though the responses are waiting in the queue.
I think the problem is all the REGISTER refresh timers expire at the
same time, and
I want to generate my own starting CSeq for my REGSITERs, but the
SIPTAG_CSEQ_STR() tag does not seem to set the CSeq in the packet.
I think I have traced this down in the code.
In nua_creq_msg(), if there is no leg yet, it calls
nta_msg_request_complete() with the default handle leg. This update
hi,
this is just a C question
i never saw anything like this:
cli_t cli[1] = sizeof(cli);
please can anyone tell me what it means? I've looked aournd for this but haven't found anything. I think It's like a malloc or something similar (??).
thanks
Luca
__
I haven't found a complete example using the user agent-level (nua) of
the API, if there is such I would appreciate an URL to it.
The space in the code snippet I sent is not in the code, so it can't be the problem. (I edited the code in the mail editor before I sent it, must have slipped on the sp
On 6/8/06, nawab hasan <[EMAIL PROTECTED]> wrote:
> Does Sofia SIP parser identify the Voicemail or IVR specific SIP URI params
> like target-param and cause-param.
Nope. All the URI params are just stored as a single sting in
url_params field. There are functions like url_param(),
url_has_param()
On 6/8/06, Colin Whittaker <[EMAIL PROTECTED]> wrote:
> Using nta and tport debugging, I have confirmed my hypothesis.
>
> In my application, I start all channels simultaneously.
> In this test, my application calls 48 nua_register() commands before
> calling su_root_step().
> What I see is all 48
On 6/9/06, Tiago Loureiro <[EMAIL PROTECTED]> wrote:
> Thanks for answering my doubts... however, I the problem remains. You have
> tried exactly the code I have supplied on the mailing list? Because running
> that same test, memory starts climbing up to 40 megs and further... ( by the
> way, by s
I had a quick peek on the code and I'm not sure what is the reason ofthe memory consumption. Please note that the SIP stack keeps a copy of
each BYE request and response for 30 seconds, just in case there isretransmission. The copy of request could be freed, but in expense ofcopying more data when
