[asterisk-dev] Asterisk tries to transcode vp8 <> g729

Dennis Buteyn dennis.buteyn at xorcom.com
Mon Jul 6 03:16:02 CDT 2020


We found that during testing that when codecs are defined in the order 
"vp8,g729" the call is dropped. When codecs are instead defined in the 
order "g729,vp8" the call succeeds. For our systems we will make sure 
codecs are always in "audio,video" order because that seems to work. It 
does smell like there's a bug somewhere but I'm not sure what... Think 
it is worth reporting or we made a mistake and docs declare we 
configured chan_sip the wrong way?

Dennis Buteyn
Xorcom Ltd

On 7/5/20 9:10 PM, Joshua C. Colp wrote:
> On Sun, Jul 5, 2020 at 12:04 PM Dennis Buteyn 
> <dennis.buteyn at xorcom.com <mailto:dennis.buteyn at xorcom.com>> wrote:
>
>     Any good reason why Asterisk thinks transcoding audio to video and
>     vice
>     versa is possible?
>
>     Capabilities: us - (vp8|g729), peer -
>     audio=(g729)/video=(vp8)/text=(nothing), combined - (vp8|g729)
>     Non-codec capabilities (dtmf): us - 0x1 (telephone-event|), peer -
>     0x1
>     (telephone-event|), combined - 0x1 (telephone-event|)
>     [2020-07-05 13:25:53] WARNING[2295][C-00000028]: channel.c:5850
>     set_format: Unable to find a codec translation path: (vp8) -> (g729)
>     [2020-07-05 13:25:53] WARNING[2295][C-00000028]: channel.c:5850
>     set_format: Unable to find a codec translation path: (g729) -> (vp8)
>
>
> The legacy way of storing format information combines all types 
> together, which can result in code mistakenly trying to set up such 
> translation paths. Entirely possible that there is a bug somewhere 
> where the video is not eliminated.
>
> -- 
> Joshua C. Colp
> Asterisk Technical Lead
> Sangoma Technologies
> Check us out at www.sangoma.com <http://www.sangoma.com> and 
> www.asterisk.org <http://www.asterisk.org>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20200706/a7fb3eda/attachment.html>


More information about the asterisk-dev mailing list