i noticed that sems master ivr apps cause sems to crash at least when caller puts play on hold using re-invite. the crash is not exactly predictable, but usually it requires two or four re-invites (one or two hold/unholds), before it happens.
here is an example using pps/examples/ivr_examples/play.py (i have converted debugs to infos): Mar 16 08:04:38 sip /usr/sbin/sip-proxy[11324]: INFO: Routing initial INVITE <sip:[email protected]> to Application Server at <sip:127.0.0.1:5090> Mar 16 08:04:38 sip sems[12010]: [#b4df6b70] [ivr_log, Ivr.cpp:80] INFO: Ivr-Python: Received play INVITE Mar 16 08:04:38 sip /usr/sbin/sip-proxy[11324]: INFO: Routing in-dialog ACK <sip:127.0.0.1:5090> from <sip:[email protected]> Mar 16 08:04:46 sip /usr/sbin/sip-proxy[11331]: INFO: Routing in-dialog INVITE <sip:127.0.0.1:5090> from <sip:[email protected]> Mar 16 08:04:46 sip sems[12010]: [#b4df6b70] [ivr_log, Ivr.cpp:80] INFO: Ivr-Python: Received play re-INVITE Mar 16 08:04:46 sip sems[12010]: [#b54fdb70] [freePacket, AmRtpStream.cpp:1148] ERROR: freePacket() double free: n_used = 0, idx = 13 Mar 16 08:04:46 sip /usr/sbin/sip-proxy[11331]: INFO: Routing in-dialog ACK <sip:127.0.0.1:5090> from <sip:[email protected]> Mar 16 08:04:48 sip /usr/sbin/sip-proxy[11331]: INFO: Routing in-dialog INVITE <sip:127.0.0.1:5090> from <sip:[email protected]> Mar 16 08:04:48 sip sems[12010]: [#b4df6b70] [ivr_log, Ivr.cpp:80] INFO: Ivr-Python: Received play re-INVITE Mar 16 08:04:48 sip /usr/sbin/sip-proxy[11331]: INFO: Routing in-dialog ACK <sip:127.0.0.1:5090> from <sip:[email protected]> Mar 16 08:04:48 sip kernel: [ 9050.042411] sems[12023]: segfault at b4cf6000 ip 0818cdbd sp b54fb700 error 6 in sems[8048000+1ac000] notice in above also error message about double free. i do have not tried with other than ivr apps so i don't know if this is an ivr issue or something more general. gdb dump is below. let me know if you need more info. it would be nice to get this fixed. -- juha program terminated with signal 11, Segmentation fault. #0 0x0818cdbd in ResampleSincMono::resample (this=0x8faf018, dst=0xb5c65000, rate=4, num_samples=2068) at resample.cpp:288 288 *dst = (short)samp; dst++; (gdb) where #0 0x0818cdbd in ResampleSincMono::resample (this=0x8faf018, dst=0xb5c65000, rate=4, num_samples=2068) at resample.cpp:288 #1 0x08096dbd in AmInternalResamplerState::resample (this=0x8faf588, samples= 0xb5c62688 "", s=3688, ratio=4) at AmAudio.cpp:243 #2 0x0809663a in AmAudio::resample (this=0xb5c6267c, rstate=..., buffer= 0xb5c62688 "", s=3688, input_sample_rate=8000, output_sample_rate=32000) at AmAudio.cpp:466 #3 0x080977ea in AmAudio::resampleOutput (this=0xb5c6267c, buffer= 0xb5c62688 "", s=3688, input_sample_rate=8000, output_sample_rate=32000) at AmAudio.cpp:461 #4 0x08116aa8 in AmRtpAudio::get (this=0xb5c41008, system_ts=3681280, buffer= 0x8f8cf7c "\336\375\023\376B\376b\376l\376a\376C\376\035\376\367\375\334\375\317\375\317\375\325\375\333\375\330\375\305\375\241\375u\375I\375'\375\033\375+\375U\375\225\375\341\375\061\376{\376\275\376\364\376$\377P\377~\377\260\377\351\377$", output_sample_rate=32000, nb_samples=640) at AmRtpAudio.cpp:263 #5 0x08138f0f in AmSession::readStreams (this=0x8f98780, ts=3681280, buffer= 0x8f8cf7c "\336\375\023\376B\376b\376l\376a\376C\376\035\376\367\375\334\375\317\375\317\375\325\375\333\375\330\375\305\375\241\375u\375I\375'\375\033\375+\375U\375\225\375\341\375\061\376{\376\275\376\364\376$\377P\377~\377\260\377\351\377$") at AmSession.cpp:1401 #6 0x080efdff in AmMediaProcessorThread::processAudio (this=0x8f8cea0, ts=3681280) at AmMediaProcessor.cpp:284 #7 0x080f0085 in AmMediaProcessorThread::run (this=0x8f8cea0) at AmMediaProcessor.cpp:256 #8 0x08161ea6 in AmThread::_start (_t=0x8f8cea0) at AmThread.cpp:66 #9 0xb76f7955 in start_thread (arg=0xb5483b70) at pthread_create.c:300 #10 0xb75501de in clone () at ../sysdeps/unix/sysv/linux/i386/clone.S:130 _______________________________________________ Semsdev mailing list [email protected] http://lists.iptel.org/mailman/listinfo/semsdev
