1270230 - WebRTC Offer SDP should include tmmbr by default and answer offers with tmmbr

Status: REOPENED

(Core :: WebRTC: Signaling, defect, P3)

Description

[Randell Jesup [:jesup] (needinfo me)]

+++ This bug was initially created as a clone of Bug #1150609 +++

Re-enable TMMBR support once tested (especially against ciscospark)

See media.navigator.video.use_tmmbr (note: not in all.js, a hidden pref)

Comment 1

[Michael Froman [:mjf]]

Do we want to add this to all.js?

Comment 2

[Randell Jesup [:jesup] (needinfo me)]

Probably should add to all.js

Comment 3

[Michael Froman [:mjf]]

Attachment: MozReview Request: Bug 1270230 - Add media.navigator.video.use_tmmbr to all.js and pref on, r=jesup

Review commit: https://reviewboard.mozilla.org/r/55148/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/55148/

Comment 4

[Randell Jesup [:jesup] (needinfo me)]

Comment on attachment 8756401 [details]
MozReview Request: Bug 1270230 - Add media.navigator.video.use_tmmbr to all.js and pref on, r=jesup

https://reviewboard.mozilla.org/r/55148/#review51806

r+ for patch, but let's do some smoke testing before landing (just a bit).  (See IRC)

Comment 5

[Pulsebot]

https://hg.mozilla.org/integration/mozilla-inbound/rev/f73ab8de07a3

Comment 6

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/f73ab8de07a3

Comment 7

[wanghb89]

I create a user.js in defaults/profile/ ,but it is trivial! How to solve this Question?

Comment 8

[Michael Froman [:mjf]]

(In reply to wanghb89 from comment #7)
> I create a user.js in defaults/profile/ ,but it is trivial! How to solve
> this Question?

I'm not sure what you're trying to do, but if you're trying to turn on TMMBR, it has been temporarily turned off in the prefs by Bug 1279039.  You can change this pref by going to about:config and looking for media.navigator.video.use_tmmbr.

Comment 9

[wanghb89]

(In reply to Michael Froman [:mjf] from comment #8)
> (In reply to wanghb89 from comment #7)
> > I create a user.js in defaults/profile/ ,but it is trivial! How to solve
> > this Question?
> 
> I'm not sure what you're trying to do, but if you're trying to turn on
> TMMBR, it has been temporarily turned off in the prefs by Bug 1279039.  You
> can change this pref by going to about:config and looking for
> media.navigator.video.use_tmmbr.

the edition of my firefox is 47.0 I donnt know whether it is mozilla47. I can't find media.navigator.video.use_tmmbr in my bout:config. I write pref("media.navigator.video.use_tmmbr", true) in user.js, put it in  defaults/profile/ ,It is trivial

Comment 10

[Michael Froman [:mjf]]

(In reply to wanghb89 from comment #9)
> the edition of my firefox is 47.0 I donnt know whether it is mozilla47. 
This pref doesn't exist in 47.  It should arrive in 49.

Comment 11

[wanghb89]

(In reply to Michael Froman [:mjf] from comment #10)
> (In reply to wanghb89 from comment #9)
> > the edition of my firefox is 47.0 I donnt know whether it is mozilla47. 
> This pref doesn't exist in 47.  It should arrive in 49.
I download the latest firefox.There is no 49. Can U give me a linker?

Comment 12

[wanghb89]

(In reply to Michael Froman [:mjf] from comment #10)
> (In reply to wanghb89 from comment #9)
> > the edition of my firefox is 47.0 I donnt know whether it is mozilla47. 
> This pref doesn't exist in 47.  It should arrive in 49.
I download the latest firefox.There is no 49. Can U give me a linker?

Comment 13

[Nils Ohlmeier [:drno]]

(In reply to wanghb89 from comment #12)
> I download the latest firefox.There is no 49. Can U give me a linker?

49 is currently the DeveloperEdition, which you can download from here https://www.mozilla.org/de/firefox/developer/

Comment 14

[wanghb89]

(In reply to Nils Ohlmeier [:drno] from comment #13)
> (In reply to wanghb89 from comment #12)
> > I download the latest firefox.There is no 49. Can U give me a linker?
> 
> 49 is currently the DeveloperEdition, which you can download from here
> https://www.mozilla.org/de/firefox/developer/

How to change media.peerconnection.navigator.min_bitrate by javascript?

Comment 15

[Nils Ohlmeier [:drno]]

(In reply to wanghb89 from comment #14)
> How to change media.peerconnection.navigator.min_bitrate by javascript?

User prefs can't be changed by JS. That would be too dangerous. Add-ons can change user prefs.

Comment 16

[Randell Jesup [:jesup] (needinfo me)]

reopening, since we preffed this back off in bug 1279039 shortly after this landed.

Re-investigate what caused the freeze, and if it still happens.

Comment 17

[Bulk Bug Changes for mreavy's org]

Mass change P1->P2 to align with new Mozilla triage process

Comment 18

[Sylvestre Ledru [:Sylvestre]]

Moving to p3 because no activity for at least 1 year(s).
See https://github.com/mozilla/bug-handling/blob/master/policy/triage-bugzilla.md#how-do-you-triage for more information

Comment 19

[Sylvestre Ledru [:Sylvestre]]

Moving to p3 because no activity for at least 1 year(s).
See https://github.com/mozilla/bug-handling/blob/master/policy/triage-bugzilla.md#how-do-you-triage for more information

Comment 20

[Alex]

Hi,

our WebRTC based video conferencing middle box is supporting Chrome for a couple of years now. We're also playing around with Firefox support and the audio part works as expected.

We make use of the TMMBR feature as it is part of the WebRTC standard (see https://datatracker.ietf.org/doc/html/draft-ietf-rtcweb-rtp-usage-26#section-5.1.6 -> " WebRTC Endpoints that are sending media are REQUIRED to implement support for TMMBR messages, and MUST follow bandwidth limitations set by a TMMBR message received for their SSRC. The sending of TMMBR requests is OPTIONAL.") but noticed it's not supported by Firefox WebRTC implementation and we stumbled over this ticket here.

We also found this as being the reason why it's disabled: https://bugzilla.mozilla.org/show_bug.cgi?id=1279039

My question would be: is the bug mentioned above still an issue? If not TMMBR, how should a WebRTC peer restrict the bandwidth of the sender in case of a middle box where you forward media and you want to apply the restrictions of the receivers to the source? Any advice would be appreciated.

Best
Alex

Comment 21

[Randell Jesup [:jesup] (needinfo me)]

I don't think Cisco Spark is a thing anymore (right?) We should try re-enabling, probably in conjunction with the reporter to test it. (I presume logmein.com is the company)

Comment 22

[Alex]

Thanks for the fast reply, really appreciated. Yes, it's about LogMeIn (GoTo). I would assume for now, that we can check if TMMBR for Firefox is working in our scenarios using media.navigator.video.use_tmmbr, right?

Comment 23

[Byron Campen [:bwc]]

(In reply to Alex from comment #22)

Thanks for the fast reply, really appreciated. Yes, it's about LogMeIn (GoTo). I would assume for now, that we can check if TMMBR for Firefox is working in our scenarios using media.navigator.video.use_tmmbr, right?

I believe that should work, yes, but I find myself wondering whether the tmmbr code in libwebrtc works at all:

https://bugs.chromium.org/p/webrtc/issues/detail?id=7103

Comment 24

[Alexander Matthes]

Another Alex from Goto here. We did a check with and without enabled media.navigator.video.use_tmmbr and didn't see a difference when sending TMMBR requests except for an additional ccm parameter ccm tmmbr in the SDP offer of the Firefox. However in further tests it looks like TMMBR is working nevertheless all the time anyway.

May it be that Firefox always handles TMMBR messages although it might not have sent out a ccm tmmbr parameter (and got a respective answer) at all?

Comment 25

[Byron Campen [:bwc]]

I guess libwebrtc enables tmmbr unilaterally (which is a spec violation, but whatever).