Gang Liu | 1 Dec 2009 04:43
Picon

timer B of INVITE transaction

Hi benny, nanang
     Is there any reason not follow RFC 3261 and cancel timer B after
UAC invite transaction get provisional response at calling state?

regards
Gang

static pj_status_t tsx_on_state_calling( pjsip_transaction *tsx,
	                                         pjsip_event *event )

...

2210	            /* For provisional response, only cancel retransmit when this
2211	             * is an INVITE transaction. For non-INVITE, section 17.1.2.1
2212	             * of RFC 3261 says that:
2213	             *  - retransmit timer is set to T2
2214	             *  - timeout timer F is not deleted.
2215	             */
2216	            if (tsx->method.id == PJSIP_INVITE_METHOD) {
2217	
2218	                /* Cancel timeout timer */
2219	                pjsip_endpt_cancel_timer(tsx->endpt, &tsx->timeout_timer);
2220	
2221	            } else {
2222	                if (!tsx->is_reliable) {
2223	                    tsx->retransmit_timer.id = TIMER_ACTIVE;
2224	                    pjsip_endpt_schedule_timer(tsx->endpt,
2225	                                               &tsx->retransmit_timer,
2226	                                               &t2_timer_val);
2227	                }
(Continue reading)

amit tyagi | 1 Dec 2009 08:05
Picon

"Update Error !" on Symbian

Hi All

I have compiled the pjsip for symbian S60 3rd edition FP1.Few days ago symbian_ua_gui.sisx file was working fine on the device but today When I try to install the self sign symbian_ua_gui.sisx file into the device,
I am getting the "update error !".


I have uninstalled all the previous application from the device but still getting the same error.
I have also set the device clock time according to my system but getting the same issue.
Please suggest me some solution.

Thanks in Advance

Regards
Amit

<div><p>Hi All<br><br>I have compiled the pjsip for symbian S60 3rd edition FP1.Few days ago symbian_ua_gui.sisx file was working fine on the device but today When I try to install the self sign symbian_ua_gui.sisx file into the device,<br>
I am getting the "update error !".<br><br><br>I have uninstalled all the previous application from the device but still getting the same error.<br>I have also set the device clock time according to my system but getting the same issue.<br>
Please suggest me some solution.<br><br>Thanks in Advance<br><br>Regards <br>Amit<br><br></p></div>
Saul Ibarra Corretge | 1 Dec 2009 13:12
Favicon
Gravatar

Distorted audio with g722

Hi!

I was testing g722 codec and I'm facing some distortion when trying the ZipDX demo
(sip:wbdemo <at> conf.zipdx.com) I've gone though the bug tracked and found this issue regarding the clock
rate (http://trac.pjsip.org/repos/ticket/486) however PJSUA is telling me that I'm using 16KHz:

Intive is correct:
INVITE sip:wbdemo <at> conf.zipdx.com SIP/2.0
Via: SIP/2.0/UDP 10.10.10.2:5060;rport;branch=z9hG4bKPjbNXyvh9iXDA8cdVGfW63Wq8QEh2zMQUQ
Max-Forwards: 70
From: <sip:10.10.10.2>;tag=sk0bfd-EK92bJwKVasEXH1PKLYOKN6sO
To: sip:wbdemo <at> conf.zipdx.com
Contact: <sip:10.10.10.2:5060>
Call-ID: 2KdBxLWweGZkzXu4b5Vjv6tESipzzgMq
CSeq: 18006 INVITE
Allow: PRACK, INVITE, ACK, BYE, CANCEL, UPDATE, SUBSCRIBE, NOTIFY, REFER, MESSAGE, OPTIONS
Supported: replaces, 100rel, timer, norefersub
Session-Expires: 1800
Min-SE: 90
User-Agent: PJSUA v1.4-trunk/i386-apple-darwin10.2.0
Content-Type: application/sdp
Content-Length:   453

v=0
o=- 3468657952 3468657952 IN IP4 10.10.10.2
s=pjmedia
c=IN IP4 10.10.10.2
t=0 0
a=X-nat:0
m=audio 4002 RTP/AVP 103 102 104 113 3 0 8 9 101
a=rtcp:4003 IN IP4 10.10.10.2
a=rtpmap:103 speex/16000
a=rtpmap:102 speex/8000
a=rtpmap:104 speex/32000
a=rtpmap:113 iLBC/8000
a=fmtp:113 mode=30
a=rtpmap:3 GSM/8000
a=rtpmap:0 PCMU/8000
a=rtpmap:8 PCMA/8000
a=rtpmap:9 G722/8000
a=sendrecv
a=rtpmap:101 telephone-event/8000
a=fmtp:101 0-15

200OK also:
SIP/2.0 200 Ok
Via: SIP/2.0/UDP 10.10.10.2:5060;received=81.204.182.64;branch=z9hG4bKPjbNXyvh9iXDA8cdVGfW63Wq8QEh2zMQUQ;rport=12528
From: <sip:10.10.10.2>;tag=sk0bfd-EK92bJwKVasEXH1PKLYOKN6sO
Call-ID: 2KdBxLWweGZkzXu4b5Vjv6tESipzzgMq
CSeq: 18006 INVITE
To: sip:wbdemo <at> conf.zipdx.com;tag=telStage-39d294b6-4b1506a0
Content-Length: 205
Content-Type: application/sdp
Allow: INVITE, ACK, CANCEL, BYE, REFER, NOTIFY, OPTIONS, INFO, REGISTER, SUBSCRIBE, MESSAGE
Session-Expires: 1800;refresher=uas
Contact: <sip:76.74.151.123;transport=udp>
Server: ZipDX-3.10.4
Supported: timer

v=0
o=telStage 1781 3468657952 IN IP4 76.74.151.123
s=-
c=IN IP4 76.74.151.123
t=0 0
m=audio 12596 RTP/AVP 9 101
a=rtpmap:9 G722/8000
a=rtpmap:101 telephone-event/8000
a=fmtp:101 0-15
a=ptime:20

But when I see the call quality (dq in pjsua) I see that audio is in 16KHz:
>>> dq
 13:06:44.445    pjsua_app.c  
  [CONFIRMED] To: sip:wbdemo <at> conf.zipdx.com;tag=telStage-39d294b6-4b1506a0
    Call time: 00h:00m:52s, 1st res in 322 ms, conn in 328ms
    SRTP status: Not active Crypto-suite: (null)
    #0 G722  <at> 16KHz, sendrecv, peer=76.74.151.123:12596
       RX pt=9, stat last update: 00h:00m:00.068s ago
          total 2.6Kpkt 416.1KB (520.2KB +IP hdr)  <at> avg=64.0Kbps/80.0Kbps
          pkt loss=0 (0.0%), discrd=1 (0.0%), dup=0 (0.0%), reord=1 (0.0%)
                (msec)    min     avg     max     last    dev
          loss period:   0.000   0.000   0.000   0.000   0.000
          jitter     :   0.000   2.863  70.625   5.375   2.654
       TX pt=9, ptime=20ms, stat last update: 00h:00m:11.333s ago
          total 1.3Kpkt 222.2KB (277.8KB +IP hdr)  <at> avg 34.1Kbps/42.7Kbps
          pkt loss=10 (0.7%), dup=0 (0.0%), reorder=0 (0.0%)
                (msec)    min     avg     max     last    dev 
          loss period: 200.000 200.000 200.000 200.000   0.000
          jitter     :   2.375   2.500   2.625   2.500   0.088
      RTT msec       :   0.000   0.000   0.000   0.000   0.000

Is it just a "printing bug" or am I really talking at 16KHz? Any clue of what could be causing that distortion? 

Let me know if I can provide more information on this.

Thanks in advance,

--

-- 
Saul Ibarra Corretge
AG Projects

John Martindale | 1 Dec 2009 16:05
Favicon

Sound device access causing kernel crash on SAM9263

Hi Benny,

Sorry for the slow reply - I'm new to mailing lists and had email 
settings wrong (and/or new email server had teething trouble).

I have got past the crashing stage now - changed the alsa config to 
point at hw instead of a plug and using native toolchain to build! - but 
was wondering if you or anyone can help me get a bit further.

If I run pjsua with ec-tail=0, then do cc 0 0, I get audio, but with a 
fast 'ripple' over the top. If I run it with ec-tail=0 and --stereo, 
then the 'ripple' is a lot louder. By 'ripple' I mean a continuous 
clicking, repeated at about ~5-10Hz. This sounds a little like the 
description of bursting on the website, could this be it?
I can play back and record sound fine using samples/auddemo. Also, if I 
use samples/simpleua I can get very good audio going from the board to 
the external endpoint, but I don't get any sound back from the endpoint.

Thanks for any help,
John

Nikolay Popok | 1 Dec 2009 20:58
Picon

Buddies. Crash in v1.5

Hi guys. I'm using pjlib v1.5.
My application handles network events. When network dissapears, i unregister all accounts and delete buddies from the lib by using pjsua_buddy_del(pjsua_buddy_id).
When network appears, i reregister accounts and add buddies by pjsua_buddy_add(pjsua_buddy_config,pjsua_buddy_id). Everything was fine when i was using pjlib 1.4.5,
but in the new version there are problems.

1) After i removed buddies and unregistered accounts, I got next warning in Xcode after few seconds (in debug this is not a crash, in release i didn't try to run, just in case):
malloc: *** error for object 0x18d2200: double free

Stack:
#3    0x968d038d in free
#4    0x0009c792 in default_block_free at pool_policy_malloc.c:78
#5    0x000a1c85 in pj_pool_destroy_int at pool.c:290
#6    0x000a2360 in cpool_release_pool at pool_caching.c:234
#7    0x000a17e1 in pj_pool_release at pool_i.h:92
#8    0x000bc910 in pjsip_endpt_release_pool at sip_endpoint.c:666
#9    0x000ae776 in pjsip_publishc_destroy at publishc.c:219
#10    0x000af548 in tsx_callback at publishc.c:678
#11    0x000cf493 in mod_util_on_tsx_state at sip_util_statefull.c:78
#12    0x000cc6c4 in tsx_set_state at sip_transaction.c:1108
#13    0x000cdc23 in tsx_send_msg at sip_transaction.c:1898
#14    0x000cdfc6 in tsx_retransmit at sip_transaction.c:2045
#15    0x000ce1e4 in tsx_on_state_calling at sip_transaction.c:2148
#16    0x000cc574 in tsx_timer_callback at sip_transaction.c:1055
#17    0x000a6201 in pj_timer_heap_poll at timer.c:518
#18    0x000bc95b in pjsip_endpt_handle_events2 at sip_endpoint.c:690
#19    0x000e364e in pjsua_handle_events at pjsua_core.c:1476
#20    0x000e16ca in worker_thread at pjsua_core.c:563
#21    0x0009a303 in thread_main at os_core_unix.c:473

2) When i'm trying to add buddies back, application crashes:
#0    0x000a5698 in pop_freelist at timer.c:136
#1    0x000a5d0c in schedule_entry at timer.c:300
#2    0x000a60a2 in pj_timer_heap_schedule at timer.c:472
#3    0x000bca9b in pjsip_endpt_schedule_timer at sip_endpoint.c:759
#4    0x000cefe4 in tsx_on_state_proceeding_uac at sip_transaction.c:2811
#5    0x000cd3f9 in pjsip_tsx_recv_msg at sip_transaction.c:1627
#6    0x000cbeb2 in mod_tsx_layer_on_rx_response at sip_transaction.c:830
#7    0x000bcebe in endpt_on_rx_msg at sip_endpoint.c:927
#8    0x000c2ff6 in pjsip_tpmgr_receive_packet at sip_transport.c:1473
#9    0x000c37a2 in udp_on_read_complete at sip_transport_udp.c:164
#10    0x00097b0f in ioqueue_dispatch_read_event at ioqueue_common_abs.c:550
#11    0x00099517 in pj_ioqueue_poll at ioqueue_select.c:765
#12    0x000bc9cb in pjsip_endpt_handle_events2 at sip_endpoint.c:719
#13    0x000e364e in pjsua_handle_events at pjsua_core.c:1476
#14    0x000e16ca in worker_thread at pjsua_core.c:563
#15    0x0009a303 in thread_main at os_core_unix.c:473

In pop_freelist function:

// The freelist values in the <timer_ids_> are negative, so we need
    // to negate them to get the next freelist "pointer."
    ht->timer_ids_freelist = -ht->timer_ids[ht->timer_ids_freelist]; // <-- crashes here, ht is not NULL

3) If i dont use adding/deleting, but call pjsua_buddy_update_pres(pjsua_buddy_id) when network appears, i dont get online status for buddies in callback from the lib (status is PJSUA_BUDDY_STATUS_UNKNOWN).


Maybe i could help with additional info, just tell me what u need.

<div><p>Hi guys. I'm using pjlib v1.5.<br>My application handles network events. When network dissapears, i unregister all accounts and delete buddies from the lib by using pjsua_buddy_del(pjsua_buddy_id).<br>When network appears, i reregister accounts and add buddies by pjsua_buddy_add(pjsua_buddy_config,pjsua_buddy_id). Everything was fine when i was using pjlib 1.4.5,<br>
but in the new version there are problems.<br><br>1) After i removed buddies and unregistered accounts, I got next warning in Xcode after few seconds (in debug this is not a crash, in release i didn't try to run, just in case):<br>
malloc: *** error for object 0x18d2200: double free<br><br>Stack:<br>#3&nbsp;&nbsp;&nbsp; 0x968d038d in free<br>#4&nbsp;&nbsp;&nbsp; 0x0009c792 in default_block_free at pool_policy_malloc.c:78<br>#5&nbsp;&nbsp;&nbsp; 0x000a1c85 in pj_pool_destroy_int at pool.c:290<br>
#6&nbsp;&nbsp;&nbsp; 0x000a2360 in cpool_release_pool at pool_caching.c:234<br>#7&nbsp;&nbsp;&nbsp; 0x000a17e1 in pj_pool_release at pool_i.h:92<br>#8&nbsp;&nbsp;&nbsp; 0x000bc910 in pjsip_endpt_release_pool at sip_endpoint.c:666<br>#9&nbsp;&nbsp;&nbsp; 0x000ae776 in pjsip_publishc_destroy at publishc.c:219<br>
#10&nbsp;&nbsp;&nbsp; 0x000af548 in tsx_callback at publishc.c:678<br>#11&nbsp;&nbsp;&nbsp; 0x000cf493 in mod_util_on_tsx_state at sip_util_statefull.c:78<br>#12&nbsp;&nbsp;&nbsp; 0x000cc6c4 in tsx_set_state at sip_transaction.c:1108<br>#13&nbsp;&nbsp;&nbsp; 0x000cdc23 in tsx_send_msg at sip_transaction.c:1898<br>
#14&nbsp;&nbsp;&nbsp; 0x000cdfc6 in tsx_retransmit at sip_transaction.c:2045<br>#15&nbsp;&nbsp;&nbsp; 0x000ce1e4 in tsx_on_state_calling at sip_transaction.c:2148<br>#16&nbsp;&nbsp;&nbsp; 0x000cc574 in tsx_timer_callback at sip_transaction.c:1055<br>#17&nbsp;&nbsp;&nbsp; 0x000a6201 in pj_timer_heap_poll at timer.c:518<br>
#18&nbsp;&nbsp;&nbsp; 0x000bc95b in pjsip_endpt_handle_events2 at sip_endpoint.c:690<br>#19&nbsp;&nbsp;&nbsp; 0x000e364e in pjsua_handle_events at pjsua_core.c:1476<br>#20&nbsp;&nbsp;&nbsp; 0x000e16ca in worker_thread at pjsua_core.c:563<br>#21&nbsp;&nbsp;&nbsp; 0x0009a303 in thread_main at os_core_unix.c:473<br><br>2) When i'm trying to add buddies back, application crashes:<br>#0&nbsp;&nbsp;&nbsp; 0x000a5698 in pop_freelist at timer.c:136<br>#1&nbsp;&nbsp;&nbsp; 0x000a5d0c in schedule_entry at timer.c:300<br>#2&nbsp;&nbsp;&nbsp; 0x000a60a2 in pj_timer_heap_schedule at timer.c:472<br>
#3&nbsp;&nbsp;&nbsp; 0x000bca9b in pjsip_endpt_schedule_timer at sip_endpoint.c:759<br>#4&nbsp;&nbsp;&nbsp; 0x000cefe4 in tsx_on_state_proceeding_uac at sip_transaction.c:2811<br>#5&nbsp;&nbsp;&nbsp; 0x000cd3f9 in pjsip_tsx_recv_msg at sip_transaction.c:1627<br>#6&nbsp;&nbsp;&nbsp; 0x000cbeb2 in mod_tsx_layer_on_rx_response at sip_transaction.c:830<br>
#7&nbsp;&nbsp;&nbsp; 0x000bcebe in endpt_on_rx_msg at sip_endpoint.c:927<br>#8&nbsp;&nbsp;&nbsp; 0x000c2ff6 in pjsip_tpmgr_receive_packet at sip_transport.c:1473<br>#9&nbsp;&nbsp;&nbsp; 0x000c37a2 in udp_on_read_complete at sip_transport_udp.c:164<br>#10&nbsp;&nbsp;&nbsp; 0x00097b0f in ioqueue_dispatch_read_event at ioqueue_common_abs.c:550<br>
#11&nbsp;&nbsp;&nbsp; 0x00099517 in pj_ioqueue_poll at ioqueue_select.c:765<br>#12&nbsp;&nbsp;&nbsp; 0x000bc9cb in pjsip_endpt_handle_events2 at sip_endpoint.c:719<br>#13&nbsp;&nbsp;&nbsp; 0x000e364e in pjsua_handle_events at pjsua_core.c:1476<br>#14&nbsp;&nbsp;&nbsp; 0x000e16ca in worker_thread at pjsua_core.c:563<br>
#15&nbsp;&nbsp;&nbsp; 0x0009a303 in thread_main at os_core_unix.c:473<br><br>In pop_freelist function:<br><br>// The freelist values in the &lt;timer_ids_&gt; are negative, so we need<br>&nbsp;&nbsp;&nbsp; // to negate them to get the next freelist "pointer."<br>
&nbsp;&nbsp;&nbsp; ht-&gt;timer_ids_freelist = -ht-&gt;timer_ids[ht-&gt;timer_ids_freelist]; // &lt;-- crashes here, ht is not NULL<br><br>3) If i dont use adding/deleting, but call pjsua_buddy_update_pres(pjsua_buddy_id) when network appears, i dont get online status for buddies in callback from the lib (status is PJSUA_BUDDY_STATUS_UNKNOWN). <br><br><br>Maybe i could help with additional info, just tell me what u need.<br><br></p></div>
fcch2k | 1 Dec 2009 22:07
Picon

IPP Codecs

Hi,

Are you able to build and run successfully the IPP in WinCE?

Thanks,

fcch

================================

Hi,

That's what I'm trying to find out as well.
In the meantime check
http://lists.pjsip.org/pipermail/pjsip_lists.pjsip.org/2008-August/004513.html
and http://lists.pjsip.org/pipermail/pjsip_lists.pjsip.org/2009-February/006203.html

Good luck, and please let us know if you manage to successfuly build
it for Windows mobile.

Regards,
Elias.

From: manoj at ascenttelecom.comTo: pjsip at lists.pjsip.orgDate: Thu,
5 Feb 2009 13:25:55 +0530Subject: [pjsip] IPP Codecs

Hello,

Are IPP codec libs from Intel supposed to work on Windows mobile
phones also?...I mean can pjsip work with IPP libs and be installed on
windows mobile phones?

Regards,

Manoj

oto nag | 1 Dec 2009 23:37
Picon

Re: Surviving to IP address changes

Hi,

after the change of IP address I call pjsua_media_transports_create() and then when I make a new call, in the SDP there is a new updated IP. But If I change IP during the call and call pjsua_media_transports_create() , after a while I get an Segmentation fault.
The interesting thing is that after the new media transport is created I send reinvite, In the SDP field of INVITE is new address, and other node start sending RTP packets to a new IP. But because of the Segmentation fault I cannot get further. 

otonag

2009/11/22 Saul Ibarra Corretge <saul <at> ag-projects.com>

On Nov 21, 2009, at 11:25 AM, oto nag wrote:

> Hi,
>
> I am trying to build an mobile sip client (in python for now) and I have same problems as written on the page you posted. If do call.reinvite() in SDP there is still an od IP.
>
> For IP change detection I made an another thread in witch I monitor actual IP address (many times per second) and if change is detected I can execute a proper actions.. but question for me is, what should be proper actions. And if these actions can be achieved with pjsua.
>

You'll nedd to restart the transport and the accounts... and if you're using pjsua I think you'll also need to restart the whole pjsua.

As you said we can do it from outside, but it'd be better if pjsip itself was able to recover from such a situation. :)



--
Saul Ibarra Corretge
AG Projects





_______________________________________________
Visit our blog: http://blog.pjsip.org

pjsip mailing list
pjsip <at> lists.pjsip.org
http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org

<div>
<div>Hi,<br>
</div>
<div><br></div>
<div>after the change of IP address I call pjsua_media_transports_create() and then when I make a new call, in the SDP there is a new updated IP. But If I change IP during the call and call pjsua_media_transports_create() , after a while I get an Segmentation fault.<br>
The interesting thing is that after the new media transport is created I send reinvite, In the SDP field of INVITE is new address, and other node start sending RTP packets to a new IP. But because of the Segmentation fault I cannot get further.&nbsp;<br>
</div>
<div><br></div>
<div>otonag</div>
<br><div class="gmail_quote">2009/11/22 Saul Ibarra Corretge <span dir="ltr">&lt;<a href="mailto:saul <at> ag-projects.com">saul <at> ag-projects.com</a>&gt;</span><br><blockquote class="gmail_quote">
<div class="im">
<br>
On Nov 21, 2009, at 11:25 AM, oto nag wrote:<br><br>
&gt; Hi,<br>
&gt;<br>
&gt; I am trying to build an mobile sip client (in python for now) and I have same problems as written on the page you posted. If do call.reinvite() in SDP there is still an od IP.<br>
&gt;<br>
&gt; For IP change detection I made an another thread in witch I monitor actual IP address (many times per second) and if change is detected I can execute a proper actions.. but question for me is, what should be proper actions. And if these actions can be achieved with pjsua.<br>

&gt;<br><br>
</div>You'll nedd to restart the transport and the accounts... and if you're using pjsua I think you'll also need to restart the whole pjsua.<br><br>
As you said we can do it from outside, but it'd be better if pjsip itself was able to recover from such a situation. :)<br><br><br><br>
--<br><div><div class="h5">Saul Ibarra Corretge<br>
AG Projects<br><br><br><br><br><br>
_______________________________________________<br>
Visit our blog: <a href="http://blog.pjsip.org" target="_blank">http://blog.pjsip.org</a><br><br>
pjsip mailing list<br><a href="mailto:pjsip <at> lists.pjsip.org">pjsip <at> lists.pjsip.org</a><br><a href="http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org" target="_blank">http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org</a><br>
</div></div>
</blockquote>
</div>
<br>
</div>
nir elkayam | 2 Dec 2009 00:49
Picon

unsubscribe sent on diffrent port the subscribe

hi,

attached the log file,
u can see that the subscribe is sent to port 30101 but the unsubscribe is sent to 30100..?!

nir
Attachment (symbian_ua.log): application/octet-stream, 71 KiB
<div><div dir="ltr">hi,<br clear="all"><br>attached the log file,<br>u can see that the subscribe is sent to port 30101 but the unsubscribe is sent to 30100..?!<br><br>nir<br>
</div></div>
Jens Jorgensen | 2 Dec 2009 01:25

Re: Distorted audio with g722

You are really "talking" at 16kHz. Please see
http://tools.ietf.org/html/rfc3551#page-14 for an explanation. To
paraphrase: the RTP clock rate for G.722 was incorrectly specified in an
old RFC as 8000. However since it has been around a long time the RFC
has officially /kept/ this error around for compatibility reasons. Cute huh?

As to the distortion problems, unfortunately I have no ideas for you. :-(

Saul Ibarra Corretge wrote:
> Hi!
>
> I was testing g722 codec and I'm facing some distortion when trying the ZipDX demo
(sip:wbdemo <at> conf.zipdx.com) I've gone though the bug tracked and found this issue regarding the clock
rate (http://trac.pjsip.org/repos/ticket/486) however PJSUA is telling me that I'm using 16KHz:
>
> Intive is correct:
> INVITE sip:wbdemo <at> conf.zipdx.com SIP/2.0
> Via: SIP/2.0/UDP 10.10.10.2:5060;rport;branch=z9hG4bKPjbNXyvh9iXDA8cdVGfW63Wq8QEh2zMQUQ
> Max-Forwards: 70
> From: <sip:10.10.10.2>;tag=sk0bfd-EK92bJwKVasEXH1PKLYOKN6sO
> To: sip:wbdemo <at> conf.zipdx.com
> Contact: <sip:10.10.10.2:5060>
> Call-ID: 2KdBxLWweGZkzXu4b5Vjv6tESipzzgMq
> CSeq: 18006 INVITE
> Allow: PRACK, INVITE, ACK, BYE, CANCEL, UPDATE, SUBSCRIBE, NOTIFY, REFER, MESSAGE, OPTIONS
> Supported: replaces, 100rel, timer, norefersub
> Session-Expires: 1800
> Min-SE: 90
> User-Agent: PJSUA v1.4-trunk/i386-apple-darwin10.2.0
> Content-Type: application/sdp
> Content-Length:   453
>
> v=0
> o=- 3468657952 3468657952 IN IP4 10.10.10.2
> s=pjmedia
> c=IN IP4 10.10.10.2
> t=0 0
> a=X-nat:0
> m=audio 4002 RTP/AVP 103 102 104 113 3 0 8 9 101
> a=rtcp:4003 IN IP4 10.10.10.2
> a=rtpmap:103 speex/16000
> a=rtpmap:102 speex/8000
> a=rtpmap:104 speex/32000
> a=rtpmap:113 iLBC/8000
> a=fmtp:113 mode=30
> a=rtpmap:3 GSM/8000
> a=rtpmap:0 PCMU/8000
> a=rtpmap:8 PCMA/8000
> a=rtpmap:9 G722/8000
> a=sendrecv
> a=rtpmap:101 telephone-event/8000
> a=fmtp:101 0-15
>
> 200OK also:
> SIP/2.0 200 Ok
> Via: SIP/2.0/UDP 10.10.10.2:5060;received=81.204.182.64;branch=z9hG4bKPjbNXyvh9iXDA8cdVGfW63Wq8QEh2zMQUQ;rport=12528
> From: <sip:10.10.10.2>;tag=sk0bfd-EK92bJwKVasEXH1PKLYOKN6sO
> Call-ID: 2KdBxLWweGZkzXu4b5Vjv6tESipzzgMq
> CSeq: 18006 INVITE
> To: sip:wbdemo <at> conf.zipdx.com;tag=telStage-39d294b6-4b1506a0
> Content-Length: 205
> Content-Type: application/sdp
> Allow: INVITE, ACK, CANCEL, BYE, REFER, NOTIFY, OPTIONS, INFO, REGISTER, SUBSCRIBE, MESSAGE
> Session-Expires: 1800;refresher=uas
> Contact: <sip:76.74.151.123;transport=udp>
> Server: ZipDX-3.10.4
> Supported: timer
>
> v=0
> o=telStage 1781 3468657952 IN IP4 76.74.151.123
> s=-
> c=IN IP4 76.74.151.123
> t=0 0
> m=audio 12596 RTP/AVP 9 101
> a=rtpmap:9 G722/8000
> a=rtpmap:101 telephone-event/8000
> a=fmtp:101 0-15
> a=ptime:20
>
> But when I see the call quality (dq in pjsua) I see that audio is in 16KHz:
>   
>>>> dq
>>>>         
>  13:06:44.445    pjsua_app.c  
>   [CONFIRMED] To: sip:wbdemo <at> conf.zipdx.com;tag=telStage-39d294b6-4b1506a0
>     Call time: 00h:00m:52s, 1st res in 322 ms, conn in 328ms
>     SRTP status: Not active Crypto-suite: (null)
>     #0 G722  <at> 16KHz, sendrecv, peer=76.74.151.123:12596
>        RX pt=9, stat last update: 00h:00m:00.068s ago
>           total 2.6Kpkt 416.1KB (520.2KB +IP hdr)  <at> avg=64.0Kbps/80.0Kbps
>           pkt loss=0 (0.0%), discrd=1 (0.0%), dup=0 (0.0%), reord=1 (0.0%)
>                 (msec)    min     avg     max     last    dev
>           loss period:   0.000   0.000   0.000   0.000   0.000
>           jitter     :   0.000   2.863  70.625   5.375   2.654
>        TX pt=9, ptime=20ms, stat last update: 00h:00m:11.333s ago
>           total 1.3Kpkt 222.2KB (277.8KB +IP hdr)  <at> avg 34.1Kbps/42.7Kbps
>           pkt loss=10 (0.7%), dup=0 (0.0%), reorder=0 (0.0%)
>                 (msec)    min     avg     max     last    dev 
>           loss period: 200.000 200.000 200.000 200.000   0.000
>           jitter     :   2.375   2.500   2.625   2.500   0.088
>       RTT msec       :   0.000   0.000   0.000   0.000   0.000
>
> Is it just a "printing bug" or am I really talking at 16KHz? Any clue of what could be causing that
distortion? 
>
> Let me know if I can provide more information on this.
>
>
> Thanks in advance,
>
>   

--

-- 
Jens B. Jorgensen
jbj1 <at> ultraemail.net

Jens Jorgensen | 2 Dec 2009 01:40

Re: Buddies. Crash in v1.5

Ahhh, I think someone just posted a patch for what sounds like your
problem yesterday or the day before. Check the archives.

Nikolay Popok wrote:
> Hi guys. I'm using pjlib v1.5.
> My application handles network events. When network dissapears, i
> unregister all accounts and delete buddies from the lib by using
> pjsua_buddy_del(pjsua_buddy_id).
> When network appears, i reregister accounts and add buddies by
> pjsua_buddy_add(pjsua_buddy_config,pjsua_buddy_id). Everything was
> fine when i was using pjlib 1.4.5,
> but in the new version there are problems.
>
> 1) After i removed buddies and unregistered accounts, I got next
> warning in Xcode after few seconds (in debug this is not a crash, in
> release i didn't try to run, just in case):
> malloc: *** error for object 0x18d2200: double free
>
> Stack:
> #3    0x968d038d in free
> #4    0x0009c792 in default_block_free at pool_policy_malloc.c:78
> #5    0x000a1c85 in pj_pool_destroy_int at pool.c:290
> #6    0x000a2360 in cpool_release_pool at pool_caching.c:234
> #7    0x000a17e1 in pj_pool_release at pool_i.h:92
> #8    0x000bc910 in pjsip_endpt_release_pool at sip_endpoint.c:666
> #9    0x000ae776 in pjsip_publishc_destroy at publishc.c:219
> #10    0x000af548 in tsx_callback at publishc.c:678
> #11    0x000cf493 in mod_util_on_tsx_state at sip_util_statefull.c:78
> #12    0x000cc6c4 in tsx_set_state at sip_transaction.c:1108
> #13    0x000cdc23 in tsx_send_msg at sip_transaction.c:1898
> #14    0x000cdfc6 in tsx_retransmit at sip_transaction.c:2045
> #15    0x000ce1e4 in tsx_on_state_calling at sip_transaction.c:2148
> #16    0x000cc574 in tsx_timer_callback at sip_transaction.c:1055
> #17    0x000a6201 in pj_timer_heap_poll at timer.c:518
> #18    0x000bc95b in pjsip_endpt_handle_events2 at sip_endpoint.c:690
> #19    0x000e364e in pjsua_handle_events at pjsua_core.c:1476
> #20    0x000e16ca in worker_thread at pjsua_core.c:563
> #21    0x0009a303 in thread_main at os_core_unix.c:473
>
> 2) When i'm trying to add buddies back, application crashes:
> #0    0x000a5698 in pop_freelist at timer.c:136
> #1    0x000a5d0c in schedule_entry at timer.c:300
> #2    0x000a60a2 in pj_timer_heap_schedule at timer.c:472
> #3    0x000bca9b in pjsip_endpt_schedule_timer at sip_endpoint.c:759
> #4    0x000cefe4 in tsx_on_state_proceeding_uac at sip_transaction.c:2811
> #5    0x000cd3f9 in pjsip_tsx_recv_msg at sip_transaction.c:1627
> #6    0x000cbeb2 in mod_tsx_layer_on_rx_response at sip_transaction.c:830
> #7    0x000bcebe in endpt_on_rx_msg at sip_endpoint.c:927
> #8    0x000c2ff6 in pjsip_tpmgr_receive_packet at sip_transport.c:1473
> #9    0x000c37a2 in udp_on_read_complete at sip_transport_udp.c:164
> #10    0x00097b0f in ioqueue_dispatch_read_event at
> ioqueue_common_abs.c:550
> #11    0x00099517 in pj_ioqueue_poll at ioqueue_select.c:765
> #12    0x000bc9cb in pjsip_endpt_handle_events2 at sip_endpoint.c:719
> #13    0x000e364e in pjsua_handle_events at pjsua_core.c:1476
> #14    0x000e16ca in worker_thread at pjsua_core.c:563
> #15    0x0009a303 in thread_main at os_core_unix.c:473
>
> In pop_freelist function:
>
> // The freelist values in the <timer_ids_≥ are negative, so we need
>     // to negate them to get the next freelist "pointer."
>     ht->timer_ids_freelist = -ht-≥timer_ids[ht->timer_ids_freelist];
> // <-- crashes here, ht is not NULL
>
> 3) If i dont use adding/deleting, but call
> pjsua_buddy_update_pres(pjsua_buddy_id) when network appears, i dont
> get online status for buddies in callback from the lib (status is
> PJSUA_BUDDY_STATUS_UNKNOWN).
>
>
> Maybe i could help with additional info, just tell me what u need.
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Visit our blog: http://blog.pjsip.org
>
> pjsip mailing list
> pjsip <at> lists.pjsip.org
> http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org
>   

--

-- 
Jens B. Jorgensen
jbj1 <at> ultraemail.net


Gmane