Asterisk 13.2.0 Video Issues
I recently compiled asterisk 13.2.0 on an RK3288 , I am suspecting problems with the format H264, Asterisk 12.8.1 compiled on the same hardware is behaving very well for the same format H264
Problem of asterisk 12.8.1 is IAX2 trunk bad voice quality.
Could someone investigate the problem of Asterisk 13 with video support on H264 ?
Thank you.
12 thoughts on - Asterisk 13.2.0 Video Issues
There’s no where near enough information in your e-mail to give someone an indication on where to start.
What channels are involved? What are their configurations? What formats are negotiated on the channels? What symptoms do you see? What does the CLI show, both when active calls are running and for a ‘core show channel’ for the involved parties?
Thank you, I needed a starting point to start my post.
1. Asterisk 12.8.1 (IAX2 voice issues) no video issues. Voice issues on IAX2 Trunks, All extensions are SIP. The IAX2 trunks on Asterisk 12.8.1 produces only one error out of : iax2
set debug trunk on
[2015-03-10 16:28:42] WARNING[9614][C-0000000b]: chan_iax2.c:1793
compress_subclass: Can’t compress subclass 2097217
There’s no where near enough information in your e-mail to give someone an indication on where to start.
What channels are involved? What are their configurations? What formats are negotiated on the channels? What symptoms do you see? What does the CLI
show, both when active calls are running and for a ‘core show channel’ for the involved parties?
Should I unload or rename the res_format_attr_h264.so H.264 Format Attribute Module The asterisk server 13.2.0 does not break anymore upon calls towards GXV3175
grandstream, however only downstream video displayed on the GXV3175 is very slow (1 frame per 10 seconds)
This problem only concerns GXV3175 for the moment (with the res_format_attr_h264.so removed). (GXV3175 version Hardware : 1.4A , program version: 1.0.3.76 and CPE version 1.0.1.32)
Any idea why ? and how could this be fixed ?
—–Original Message—
Both Asterisk 1.6.2 and Asterisk 12 no longer receive bug fixes, so I’m going to skip past this issue.
Asterisk crashing is a bug. That’s a bad thing. Please get a backtrace
[1] and file an issue on the issue tracker [2]. A pcap of the message traffic would also be very helpful.
I’m going to assume “Asterisk stops” means it crashed as well. If you’d like to get a backtrace for that as well and attach it to the same issue, that would be helpful – it may be the same problem that you see with the Bria phone, or it may be something else.
Well, that’s good. 🙂
Since there are some that work fine, and some that don’t, the trick is going to be knowing:
(1) How the SIP peers (or PJSIP endpoints) are configured
(2) How the phones are negotiating media with Asterisk
Both your SIP configuration as well as a DEBUG log – generated with trace logging, showing the negotiation [3] – will be needed to figure out what is occurring.
[1] https://wiki.asterisk.org/wiki/display/AST/Getting+a+Backtrace
[2] https://issues.asterisk.org/jira/
[3] https://wiki.asterisk.org/wiki/display/AST/Collecting+Debug+Information
I will rebuild my asterisk with the options enabled ONT_OPTIMIZE and BETTER_BACKTRACES
Then I will create the traces and post them as per your recommendations.
—–Original Message—
Hello Matthew,
I have compiled Asterisk 13.2 with the following compiler Flags enabled:
DON’T_OPTIMIZE
DEBUG THREADS
BETTER_BACKTRACES
My asterisk is running with the asterisk_script:
root 24048 39.4 2.4 128564 50640 pts/1 Sl 00:02 2:21
/usr/sbin/asterisk -f -vvvg -c
core show locks
========================================================================= 13.2.0 (): (times locked)
=== Currently Held Locks
===========================================================================
========================================================================
When my asterisk crashes there is no file called core.
The results of gdb -se “asterisk” -ex “bt full” -ex “thread apply all bt”
–batch -c core > /tmp/backtrace.txt
/usr/src/asterisk-13.2.0/core: No such file or directory. No stack.
What could be the problem ?
Best regards Toufic
—–Original Message—
(1) Asterisk only generates a core file if started with the ‘-g’ option
(2) Your core file may not be located in the directory that you are running gdb from. You will need to find where the core file was located – this is typically determined by
/proc/sys/kernel/core_pattern
I see that my asterisk is started with the -g option, the core file I cannot find on my system (find / -name core*)
—–Original Message—
I would suspect one of the following:
(1) Asterisk is not actually crashing.
(2) Something is deleting the core files.
(3) The core files are hiding really, really well.
Either way, if you can’t get a backtrace, there isn’t much we can do to help with that problem.
I can assure you that asterisk is crashing, as when I try to reconnect I see it reloading again. Could be that something is deleting the core ! is there a way to find the path to where the core files are stored?
My system is Lubuntu , Linux #41 SMP PREEMPT Tue Nov 11 16:35:58 CST 2014
armv7l armv7l armv7l GNU/Linux Operating system Ubuntu Linux 14.04.1
–
If you take a look at the safe_asterisk shell script, usually located at
/usr/sbin/safe_asterisk (for CentOS at least), you’ll be able to find where the core files are located. If it’s not located there, then you’ll need to look at the Asterisk init script for the scripts location. I hope this helps. Regards;
John
—–Original Message—
Hello Matthew,
The asterisk crashing issue was solved with the Asterisk 13.3.0, now video calls are okay between all devices. The only issue left is with the Grandstream GXV3175 where video is still very slow (downstream), it shows on the LCD 1 frame every few seconds.
Hope this helps and should someone has a suggestion on how to solve the GXV3175 video would be great.
Best regards Toufic
—–Original Message—