[jitsi-dev] Questions about Jigasi logs and dial-out


#1

Hi Jitsi Team!

I'm testing Jigasi with Jitsi Meet from the unstable branch.

When checking the log it seems that Jigasi has been able to register:

···

----------------------------------------------------------------
2018-04-13 19:59:44.873 INFO: [44]
org.jitsi.jigasi.AbstractGateway.registrationStateChanged().120 REG
STATE CHANGE ProtocolProviderServiceSipImpl(12345678@1.2.3.4 (SIP)) ->
RegistrationStateChangeEvent[ oldState=Unregistered;
newState=RegistrationState=Registering; reasonCode=-1; reason=null]
2018-04-13 19:59:44.953 INFO: [44]
impl.protocol.sip.ProxyRouter.getNextHop().167 Outbound proxy mode,
using proxy 1.2.3.4:5060/UDP as hop instead of an address resolved by
the SIP router
(...)
2018-04-13 19:59:45.206 INFO: [51]
org.jitsi.jigasi.AbstractGateway.registrationStateChanged().120 REG
STATE CHANGE ProtocolProviderServiceSipImpl(12345678@1.2.3.4 (SIP)) ->
RegistrationStateChangeEvent[ oldState=Registering;
newState=RegistrationState=Registered; reasonCode=-1; reason=null]
----------------------------------------------------------------

But then I see this pair of entries that are repeated every 25 seconds:

----------------------------------------------------------------
2018-04-13 22:36:25.253 INFO: [53]
impl.protocol.sip.ProxyRouter.getNextHop().167 Outbound proxy mode,
using proxy 1.2.3.4:5060/UDP as hop instead of an address resolved by
the SIP router
2018-04-13 22:36:25.319 INFO: [483]
impl.protocol.sip.SipLogger.logInfo().196 Info from the JAIN-SIP stack:
Setting SIPMessage peerPacketSource to: /1.2.3.4:5060
----------------------------------------------------------------

Is this the expected behavior?

On the other hand, to dial-out I have to go to the plus sign? In the
stable branch I remember that a window with the country codes was shown
and you had to configure the countries in the file specified by the
'dialOutCodesUrl' variable in config.js. In the version for the unstable
branch is it disabled to allow free dialing by manually including the
country code or not?

Thanks in advance.

Kind regards,
Daniel


#2

The log about the next hop is normal. I've increased the level of this statement for troubleshooting and didn't expect to see it this often. IIRC it's changed back meanwhile in Jitsi Desktop, but the Jigasi dependencies haven't been updated since.

Ingo

Freundliche Grüsse,
Ingo Bauersachs

-- sent from my mobile

···

On 14 Apr 2018, at 00:58, Daniel Bareiro <daniel-listas@gmx.net> wrote:

getNextHop


#3

Hi, Ingo.

The log about the next hop is normal. I've increased the level of this
statement for troubleshooting and didn't expect to see it this often.
IIRC it's changed back meanwhile in Jitsi Desktop, but the Jigasi
dependencies haven't been updated since.

Thanks for your reply. Yes, I basically had doubts about this because I
have hundreds of these entries in the Jigasi log. Beyond this, outgoing
calls seem to be working.

For incoming SIP calls with Jigasi, have you currently implemented video
reception (or are you planning to implement it in the future)?

Kind regards,
Daniel

···

On 14/04/18 04:59, Ingo Bauersachs wrote:


#4

Hi,

Hi, Ingo.

> The log about the next hop is normal. I've increased the level of this
> statement for troubleshooting and didn't expect to see it this often.
> IIRC it's changed back meanwhile in Jitsi Desktop, but the Jigasi
> dependencies haven't been updated since.

Thanks for your reply. Yes, I basically had doubts about this because I
have hundreds of these entries in the Jigasi log. Beyond this, outgoing
calls seem to be working.

We will be working on updating jigasi soon.

For incoming SIP calls with Jigasi, have you currently implemented video
reception (or are you planning to implement it in the future)?

Video for jigasi is not supported and there is no plan to support it. We
are succesfully using jibri with sip video devices.
We are working/updating that functionality at the moment, as jibri was
recently rewriten to more stable and easy to deploy version.

Regards
damencho

···

On Sat, Apr 14, 2018 at 1:25 PM, Daniel Bareiro <daniel-listas@gmx.net> wrote:

On 14/04/18 04:59, Ingo Bauersachs wrote:

Kind regards,
Daniel

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


#5

Hi,

Hi, Damian.

> Thanks for your reply. Yes, I basically had doubts about this because I
> have hundreds of these entries in the Jigasi log. Beyond this, outgoing
> calls seem to be working.

We will be working on updating jigasi soon.

Thank you for this.

> For incoming SIP calls with Jigasi, have you currently implemented video
> reception (or are you planning to implement it in the future)?

Video for jigasi is not supported and there is no plan to support it. We
are succesfully using jibri with sip video devices.

Thanks for confirming this. Do you have any documentation to configure
and test with sip video devices using Jibri?

We are working/updating that functionality at the moment, as jibri was
recently rewriten to more stable and easy to deploy version.

Yes. I've seen Brian's email telling about it. Thank you for the work
done in the new version of Jibri. Much appreciated.

I have noticed that in the unstable repo there is a Jibri package. I
would like to know if it is the new version has been mentioned.

In the Github documentation I read that Jibri is intended to be run on a
separate machine. Is there any reason why it is not advisable to run it
on the same machine as Jitsi Meet?

Thanks for your time.

Kind regards,
Daniel

···

On 16/04/18 16:19, Damian Minkov wrote:


#6

Hi again,

> Hi,

Hi, Damian.

> > Thanks for your reply. Yes, I basically had doubts about this because I
> > have hundreds of these entries in the Jigasi log. Beyond this, outgoing
> > calls seem to be working.

> We will be working on updating jigasi soon.

Thank you for this.

> > For incoming SIP calls with Jigasi, have you currently implemented
video
> > reception (or are you planning to implement it in the future)?

> Video for jigasi is not supported and there is no plan to support it. We
> are succesfully using jibri with sip video devices.

Thanks for confirming this. Do you have any documentation to configure
and test with sip video devices using Jibri?

This is comming, Brian is working on it.

> We are working/updating that functionality at the moment, as jibri was
> recently rewriten to more stable and easy to deploy version.

Yes. I've seen Brian's email telling about it. Thank you for the work
done in the new version of Jibri. Much appreciated.

I have noticed that in the unstable repo there is a Jibri package. I
would like to know if it is the new version has been mentioned.

Yes, this is the one. The previous one did not have debian package.

In the Github documentation I read that Jibri is intended to be run on a
separate machine. Is there any reason why it is not advisable to run it
on the same machine as Jitsi Meet?

Well there is no point of running it there, any way if you want to support
simultenious sessions, multiple sip device, basically you need on jibri per
device.
But if you have one device, I suppose it is ok to run it on the same
machine, the thing is that it is taking a lot of cpu usage and there should
be some left for jvb.

Regards
damencho

···

On Mon, Apr 16, 2018 at 3:03 PM, Daniel Bareiro <daniel-listas@gmx.net> wrote:

On 16/04/18 16:19, Damian Minkov wrote:

Thanks for your time.

Kind regards,
Daniel

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


#7

Hi again,

Hi, Damian.

> > Video for jigasi is not supported and there is no plan to support it. We
> > are succesfully using jibri with sip video devices.

> Thanks for confirming this. Do you have any documentation to configure
> and test with sip video devices using Jibri?

This is comming, Brian is working on it.

Great! Just to make it clear, would Jibri allow video in both incoming
and outgoing calls?

> > We are working/updating that functionality at the moment, as jibri was
> > recently rewriten to more stable and easy to deploy version.

> Yes. I've seen Brian's email telling about it. Thank you for the work
> done in the new version of Jibri. Much appreciated.
>
> I have noticed that in the unstable repo there is a Jibri package. I
> would like to know if it is the new version has been mentioned.

Yes, this is the one. The previous one did not have debian package.

Perfect. Thanks for confirm.

> In the Github documentation I read that Jibri is intended to be run on a
> separate machine. Is there any reason why it is not advisable to run it
> on the same machine as Jitsi Meet?

Well there is no point of running it there, any way if you want to
support simultenious sessions, multiple sip device, basically you need
on jibri per device.

But if you have one device, I suppose it is ok to run it on the same
machine, the thing is that it is taking a lot of cpu usage and there
should be some left for jvb.

I see. My initial idea is to use Jibri to record conference rooms. So a
single instance of Jibri is able to only record one conference room at a
time (regardless of the number of participants in the conference room)?

I'm not sure if I understood what you said about "support multiple sip
devices". Suppose that from a conference room I want to call two people
using sip video devices. In that case, a single instance of Jibri can
connect those two devices to the conference room or would it need two
instances of Jibri?

Thanks for your time.

Kind regards,
Daniel

···

On 16/04/18 17:29, Damian Minkov wrote: