[jitsi-users] Trickle ICE at webrtcH4cKS


#1

Hi all,

The webrtcH4cKS blog has just published a handy Trickle ICE tutorial (http://goo.gl/VTwXm8) on NAT traversal and fast session establishment from Jitsi's Emil Ivov.

Cheers,
Yana


#2

Thanks,
this is a rather nice explanation of the NAT issues.
I'm not sure that I've fully grasped half-trickle. I guess it could be
(ICE) implementation dependent whether it works or not, right? I guess
there could be problematic timeouts.

Anyway, thanks for the link.

Regards,
Philipp

···

On Thu, 5 Dec 2013 11:23:20 +0100 Yana Stamcheva <yana@jitsi.org> wrote:

Hi all,

The webrtcH4cKS blog has just published a handy Trickle ICE tutorial
(http://goo.gl/VTwXm8) on NAT traversal and fast session
establishment from Jitsi's Emil Ivov.

Cheers,
Yana


#3

Emil is falling in love with WebRTC lately! :-p

···

On Thu, 5 Dec 2013, Yana Stamcheva wrote:

The webrtcH4cKS blog has just published a handy Trickle ICE tutorial (http://goo.gl/VTwXm8) on NAT traversal and fast session establishment from Jitsi's Emil Ivov.


#4

Hey Philipp,

···

On 5 Dec 2013 13:03, "Philipp Überbacher" <murks@tuxfamily.org> wrote:

I'm not sure that I've fully grasped half-trickle.

Well simply put it starts like regular ICE and if both sides see that their
peer has support for trickle, they switch to that.

This things work regardless of whether implementations have support for
trickle.

Emil

--sent from my mobile


#5

Hey,

···

On 12/05/2013 01:22 PM, Emil Ivov wrote:

On 5 Dec 2013 13:03, "Philipp �berbacher" <murks@tuxfamily.org> wrote:

I'm not sure that I've fully grasped half-trickle.

Well simply put it starts like regular ICE and if both sides see that their
peer has support for trickle, they switch to that.

This things work regardless of whether implementations have support for
trickle.

What I'm actually missing is a pointer to how this is supposed to work
with SIP. Or did I overlook it?

Andreas


#6

This needs more work and we'll be updating it soon but the idea is in there:

http://tools.ietf.org/html/draft-ivov-mmusic-trickle-ice-sip

Emil

···

On Thu, Dec 5, 2013 at 5:36 PM, Andreas Granig <agranig@sipwise.com> wrote:

Hey,

On 12/05/2013 01:22 PM, Emil Ivov wrote:

On 5 Dec 2013 13:03, "Philipp Überbacher" <murks@tuxfamily.org> wrote:

I'm not sure that I've fully grasped half-trickle.

Well simply put it starts like regular ICE and if both sides see that their
peer has support for trickle, they switch to that.

This things work regardless of whether implementations have support for
trickle.

What I'm actually missing is a pointer to how this is supposed to work
with SIP. Or did I overlook it?

--
https://jitsi.org


#7

By the way, somewhat off topic, but has jitsi got ICE support for SIP yet?
I remember it was supposed to be done by summer but IIRC it kept getting
put off.

Trickle-ice sounds very interesting. Will have to take a deeper look.

Thanks

···

On Thu, Dec 5, 2013 at 4:39 PM, Emil Ivov <emcho@jitsi.org> wrote:

On Thu, Dec 5, 2013 at 5:36 PM, Andreas Granig <agranig@sipwise.com> > wrote:
> Hey,
>
> On 12/05/2013 01:22 PM, Emil Ivov wrote:
>> On 5 Dec 2013 13:03, "Philipp Überbacher" <murks@tuxfamily.org> wrote:
>>> I'm not sure that I've fully grasped half-trickle.
>>
>> Well simply put it starts like regular ICE and if both sides see that
their
>> peer has support for trickle, they switch to that.
>>
>> This things work regardless of whether implementations have support for
>> trickle.
>
> What I'm actually missing is a pointer to how this is supposed to work
> with SIP. Or did I overlook it?

This needs more work and we'll be updating it soon but the idea is in
there:

http://tools.ietf.org/html/draft-ivov-mmusic-trickle-ice-sip

Emil

--
https://jitsi.org

_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev


#8

By the way, somewhat off topic, but has jitsi got ICE support for SIP yet?

Ah ... sigh. Unfortunately not yet. That took some delay (Totally my fault). It's still very much planned though.

I remember it was supposed to be done by summer but IIRC it kept getting
put off.

It didn't get put off. It just never got put on due to other (real life) things that I've needed to deal with.

Emil

···

On 05.12.13, 21:15, Privus 007 wrote:

Trickle-ice sounds very interesting. Will have to take a deeper look.

Thanks

On Thu, Dec 5, 2013 at 4:39 PM, Emil Ivov <emcho@jitsi.org > <mailto:emcho@jitsi.org>> wrote:

    On Thu, Dec 5, 2013 at 5:36 PM, Andreas Granig <agranig@sipwise.com > <mailto:agranig@sipwise.com>> wrote:
     > Hey,
     >
     > On 12/05/2013 01:22 PM, Emil Ivov wrote:
     >> On 5 Dec 2013 13:03, "Philipp Überbacher" <murks@tuxfamily.org > <mailto:murks@tuxfamily.org>> wrote:
     >>> I'm not sure that I've fully grasped half-trickle.
     >>
     >> Well simply put it starts like regular ICE and if both sides see
    that their
     >> peer has support for trickle, they switch to that.
     >>
     >> This things work regardless of whether implementations have
    support for
     >> trickle.
     >
     > What I'm actually missing is a pointer to how this is supposed to
    work
     > with SIP. Or did I overlook it?

    This needs more work and we'll be updating it soon but the idea is
    in there:

    http://tools.ietf.org/html/draft-ivov-mmusic-trickle-ice-sip

    Emil

    --
    https://jitsi.org

    _______________________________________________
    dev mailing list
    dev@jitsi.org <mailto:dev@jitsi.org>
    Unsubscribe instructions and other list options:
    http://lists.jitsi.org/mailman/listinfo/dev

_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev

--
https://jitsi.org