LiveKit Community

is this Cloud Egress or self-hosted <@U05E0KG6YQY>?

cc <@U01UB0LHTTQ>

Hey simple question for you guys, what's the easiest way to know if the local participant is subscribed to any publication?

Hello <@U02SHDC9J7N>, playground has been removed.

Our goal is to enable developers to work in their preferred development environment. Having to switch when you outgrow the playground environment is not ideal, in our opinion.
But perhaps David has more to add.

Apologies <@U02SHDC9J7N>, Playground was low traffic and given all the other things were working on, a burden to maintain. Here’s a quick and more powerful substitute though: <https://replit.com/@RussdSa/meet> which you can fork and follow the instructions in the README (also here: <https://github.com/livekit-examples/meet>) to get up-and-running.

Hi <@U05837ULPJ6>, LiveKit doesn’t support something like that out of the box, however, you could build it. One way you could do it:
1. In your application, build a simple page/component which allows someone to upload a presentation file
2. Start an RTMP-out LK Web Egress job which displays that presentation on a web page
3. Start an LK Ingress job which takes RTMP-in and feeds it back into the room

```2023/06/21 06:10:58 "level"=0 "msg"="resuming connection..." "reconnectCount"=0
2023/06/21 06:10:58 "level"=0 "msg"="received offer for subscriber"
2023/06/21 06:10:58 "level"=0 "msg"="successfully set publisher answer"
2023/06/21 06:10:58 "level"=0 "msg"="ICE connected"
2023/06/21 06:13:28 "level"=0 "msg"="resuming connection..." "reconnectCount"=0
2023/06/21 06:13:29 "level"=0 "msg"="successfully set publisher answer"
2023/06/21 06:13:29 "level"=0 "msg"="received offer for subscriber"
2023/06/21 06:13:29 "level"=0 "msg"="ICE connected"
2023/06/21 06:13:39 "level"=0 "msg"="resuming connection..." "reconnectCount"=0
2023/06/21 06:13:39 "level"=0 "msg"="received offer for subscriber"
2023/06/21 06:13:39 "msg"="could not send ICE candidates for subscriber" "error"="write tcp <http://xxx.xxx.xxx.xxx:34720-&gt;xxx.xxx.xxx.xxx:443|xxx.xxx.xxx.xxx:34720-&gt;xxx.xxx.xxx.xxx:443>: write: broken pipe"
2023/06/21 06:13:39 "msg"="could not send ICE candidates for publisher" "error"="write tcp <http://xxx.xxx.xxx.xxx:34720-&gt;xxx.xxx.xxx.xxx:443|xxx.xxx.xxx.xxx:34720-&gt;xxx.xxx.xxx.xxx:443>: write: broken pipe"
2023/06/21 06:13:54 "msg"="resume connection failed" "error"="could not connect after timeout"
2023/06/21 06:13:54 "level"=0 "msg"="restarting connection..." "reconnectCount"=1
2023/06/21 06:13:55 "level"=0 "msg"="received offer for subscriber"
2023/06/21 06:13:57 "level"=0 "msg"="ICE connected"```

I've been getting a lot more of these lately, where `resume connection failed`

```2023/06/21 06:14:37 "level"=0 "msg"="resuming connection..." "reconnectCount"=0
2023/06/21 06:14:37 "msg"="could not negotiate" "error"="ICEAgent does not exist: unable to restart ICETransport"
2023/06/21 06:14:38 "level"=0 "msg"="received offer for subscriber"
2023/06/21 06:14:40 "level"=0 "msg"="ICE connected"
2023/06/21 06:14:50 "level"=0 "msg"="resuming connection..." "reconnectCount"=0
2023/06/21 06:14:50 "msg"="could not negotiate" "error"="ICEAgent does not exist: unable to restart ICETransport"
2023/06/21 06:14:50 "level"=0 "msg"="received offer for subscriber"
2023/06/21 06:14:53 "level"=0 "msg"="ICE connected"
2023/06/21 06:15:37 "level"=0 "msg"="Leave room" "reason"="JOIN_FAILURE"```

Which version of livekit sever and server-sdk-go are you on Pat? I will check for an increase in the ICE Agent error. That looks new.

Using Cloud, and the SDK server-sdk-go v1.0.10

Looks like there is no uptick in the ICEAgent issue (I need to look at the source of when that happens, but I digress :slightly_smiling_face: )

Can you please share your project ID (you can DM me)? Will check for an up tick in resumes in your project.

I really don’t know what to do with it since I’ve tried several ways to make it work

always the `livekit_react-native:compileReleaseKotlin` complains about this …

could it be related to some caching issue, <@U01NZKTNE1H>?

maybe remove the kotlin stuff from your files if you’re not using this. I just tried a fresh build, where the template doesn’t include anything for kotlin, and works fine

<@U01NZKTNE1H> building the app for release?

just ran:
&gt; npx react-native init rnsixseven --version 0.67.5
&gt; yarn add @livekit/react-native @livekit/react-native-webrtc

but <@U01NZKTNE1H>, what should I do if I need kotlin for other libraries?

I mean I should be able to use kotlin since nowadays there are libraries with needs it, right?

you don’t need to declare the dependency if you’re not using it yourself

let me try and come back to you, is it ok?

<@U01NZKTNE1H> I’ve tried your way also and did not work

it’s insane already what am I doing wrong?

brand new app just installed the 2 packages for livekit

added the configuration from ReadMe. and compile using Android Studio in release mode the app

image.png

You can see I only have 2 packages witch are turning on to release nothing else added

```&gt; Task :app:lintVitalRelease
w: Runtime JAR files in the classpath should have the same version. These files were found in the classpath:
    {project_path}/.gradle/caches/modules-2/files-2.1/org.jetbrains.kotlin/kotlin-stdlib-common/1.6.21/5e5b55c26dbc80372a920aef60eb774b714559b8/kotlin-stdlib-common-1.6.21.jar (version 1.6)
     {project_path}/.gradle/caches/modules-2/files-2.1/org.jetbrains.kotlin/kotlin-stdlib-jdk8/1.4.10/998caa30623f73223194a8b657abd2baec4880ea/kotlin-stdlib-jdk8-1.4.10.jar (version 1.4)
     {project_path}/.gradle/caches/modules-2/files-2.1/org.jetbrains.kotlin/kotlin-stdlib-jdk7/1.4.10/30e46450b0bb3dbf43898d2f461be4a942784780/kotlin-stdlib-jdk7-1.4.10.jar (version 1.4)
     {project_path}/.gradle/caches/modules-2/files-2.1/org.jetbrains.kotlin/kotlin-stdlib/1.4.10/ea29e063d2bbe695be13e9d044dcfb0c7add398e/kotlin-stdlib-1.4.10.jar (version 1.4)
     {project_path}/.gradle/caches/modules-2/files-2.1/org.jetbrains.kotlin/kotlin-stdlib-common/1.4.10/6229be3465805c99db1142ad75e6c6ddeac0b04c/kotlin-stdlib-common-1.4.10.jar (version 1.4)
     {project_path}/.gradle/caches/modules-2/files-2.1/org.jetbrains.kotlin/kotlin-stdlib/1.6.21/11ef67f1900634fd951bad28c53ec957fabbe5b8/kotlin-stdlib-1.6.21.jar (version 1.6)```

I get this because I don’t specify the Kotlin version

can you just upload the nonworking project?

github if you can, or you can upload here, but make sure to clean the folder for node_modules and build files

<https://github.com/Mihai-github/rn-0.67.5>

it works for me. how are you building the project? I’m just running `npx react-native run-android --variant release`.

I’ve used android studio -&gt; changed the variant to release and turn project on my personal android device

I will try again to see if the error persists

could be you’ve got some other local file setting the kotlin version that didn’t get uploaded to git

try checking out your repo in a clean folder and building from there

<@U01NZKTNE1H> ok, seems like w/o any kotlin configuration works but my only concern is with this error if I need to worry about them or not

The sad part is i have the other project to upgrade it to React Native `v0.67.5`  to see if like this will work because on RN `v0.64.1`  I get the error with Unknown witch hopefully may the upgrade to a higher RN version solve this?

not sure why you get those errors, but if you’re upgrading you might as well upgrade to the latest.

you suggest to try `0.70.x`  or take small steps initially to a `0.67.5` since the upgrade will be huge

The project I am working on has more than 2y of development so there are many things in there :))

there’s a react-native upgrade helper tool that tells you what you need to do to upgrade

but usually I saw even with the tool for upgrade is not always plug and play since some other pages may need also some tweaks

Hopefully a newer version of RN can solve on my project that weird error I get with
```2023-06-18 18:42:46.665396+0300 uHubs Dev[8610:509463] [javascript] TypeError: undefined is not an object (evaluating 'r(d[8]).Room')```

<@U03GEMJGJJ3> the latter, it’s all done over WebRTC using LiveKit

what a wonderful technology and library livekit is

I will try hard to understand the code.. shouldn't be hard... I couldn't run the git repo there is an error

<@U01KZ0SCDKL> the error... layer 8 error. I don't know yet what to ask - but thank you that there is a channel here, to ask as soon as I have formulated a question.

OK I have also this error:

on npm run dev:

```hydration-error-info.js:24 The above error occurred in the &lt;Graphics&gt; component:

    at Graphics
    at Container
    at JukeBox (webpack-internal:///(app-client)/./src/components/JukeBox.tsx:19:11)
    at Container
    at Container
    at Camera (webpack-internal:///(app-client)/./src/components/Camera.tsx:11:11)
    at AnimationsProvider (webpack-internal:///(app-client)/./src/providers/animations.tsx:93:11)

Consider adding an error boundary to your tree to customize error handling behavior.
Visit <https://reactjs.org/link/error-boundaries> to learn more about error boundaries.```


on entering the room: <http://localhost:3001/room/test>

Unhandled Runtime Error
*TypeError: Class constructor _Container cannot be invoked without 'new'*
Call Stack
new Graphics
node_modules/@pixi/graphics/dist/esm/graphics.mjs (2057:0)
Graphics
node_modules/@pixi/react/dist/index.es-dev.js (479:35)
createElement
node_modules/@pixi/react/dist/index.es-dev.js (779:0)
completeWork
node_modules/@pixi/react/dist/index.es-dev.js (15082:0)
completeUnitOfWork
node_modules/@pixi/react/dist/index.es-dev.js (20208:0)
performUnitOfWork
node_modules/@pixi/react/dist/index.es-dev.js (20180:0)
workLoopSync
node_modules/@pixi/react/dist/index.es-dev.js (20078:0)
renderRootSync
node_modules/@pixi/react/dist/index.es-dev.js (20046:0)
recoverFromConcurrentError
node_modules/@pixi/react/dist/index.es-dev.js (19428:0)
performSyncWorkOnRoot
node_modules/@pixi/react/dist/index.es-dev.js (19674:0)
flushSyncCallbacks
node_modules/@pixi/react/dist/index.es-dev.js (4419:0)
eval
node_modules/@pixi/react/dist/index.es-dev.js (19226:0)

error 2:
Unhandled Runtime Error
*TypeError: Cannot read properties of undefined (reading 'texture')*
Call Stack
new AnimatedSprite
node_modules/@pixi/sprite-animated/lib/AnimatedSprite.mjs (6:40)
AnimatedSprite
node_modules/@pixi/react/dist/index.es-dev.js (589:59)
createElement
node_modules/@pixi/react/dist/index.es-dev.js (779:0)
completeWork
node_modules/@pixi/react/dist/index.es-dev.js (15082:0)
completeUnitOfWork
node_modules/@pixi/react/dist/index.es-dev.js (20208:0)
performUnitOfWork
node_modules/@pixi/react/dist/index.es-dev.js (20180:0)
workLoopSync
node_modules/@pixi/react/dist/index.es-dev.js (20078:0)
renderRootSync
node_modules/@pixi/react/dist/index.es-dev.js (20046:0)
recoverFromConcurrentError
node_modules/@pixi/react/dist/index.es-dev.js (19428:0)
performSyncWorkOnRoot
node_modules/@pixi/react/dist/index.es-dev.js (19674:0)
flushSyncCallbacks
node_modules/@pixi/react/dist/index.es-dev.js (4419:0)
eval
node_modules/@pixi/react/dist/index.es-dev.js (19226:0)

Any idea what’s happening here <@U01UEMQ8C3F> ?

<@U01KZ0SCDKL> I saw someone made an issue on the repo - it is solved. The project is using es5; but the pixi library uses es6 or so.. just need to change the tsconfig target to es6 and npm reinstall pixi with force. thx

<https://github.com/livekit/client-sdk-swift/issues/219>

This issue was reported 18 hours ago :white_check_mark:

<https://github.com/livekit/client-sdk-swift/issues/182>

<https://github.com/livekit/client-sdk-python>

This is still in WIP :slightly_smiling_face:

is client sdk different from server sdk. ? and when can we expect python SDK to be officially released.

Yes, this is different, the client SDK allows for RTC capabilities, while the server SDK allow to make requests to LiveKit services (Like creating/removing a room, or to record a track)
There is a community version of a server sdk: <https://github.com/tradablebits/livekit-server-sdk-python>

I’m not sure about when we can officially release the Python SDK, but definitely in 1-2 months

Thanks <@U022KB72NFJ>, i have been looking into server-sdk from <https://github.com/tradablebits|tradablebits> ,however i felt that sdk is incomplete, doesn't have ways to connect the bot to room.

Hey <@U050PLV6VF0>, I’m afraid I don’t have a good solution for you. Due to the nature of video/audio applications, it is inherently difficult to mock, test, and reliably execute these tests without an actual connection to the server.
The only thing (as far as I know) that simulates users is the `simulateParticipants` option here: <https://docs.livekit.io/client-sdk-js/classes/Room.html#simulateParticipants>

If you are using `@livekit/components-react` then you can also us it like this:
```&lt;LiveKitRoom
 simulateParticipants={9}
&gt;
    {...}
&lt;/LiveKitRoom&gt;```
I’m not aware of any project that has mocked API calls to LiveKit Cloud, but maybe someone in the community has built something cool.
If you want to share your setup/integration and what you want to test, we might be able to give you some more detailed feedback/tips.
I’m afraid it’s still a problem that could use some cool solutions. If you come up with something, we would love to see it! :)

Ok I will check out simulateParticipants; thank you!!

would it be possible to spin up a local LK server (with a different url and api key/secret pair) when running in the test env?

oh interesting idea, i can also look into that

you can also have a look at how we solve this in the components repo for storybook. We basically inject a dummy room context in order to make it work. The solution is however very similar to what dsa proposed already.
<https://github.com/livekit/components-js/blob/main/docs/storybook/.storybook/lk-decorators/RoomContext.tsx>

cc <@U01UB0LHTTQ> <@U02BBTRGF98> <@U02Q4J8KS85>

This is intentional to improve autoscaling and reduce cost. If you have 10 nodes each running one stream for 10 hours, it’ll be much cheaper to have them spread across 4 or 5 nodes instead

they take cpu into account and will avoid over-committing resources

ClusterID is for when you have multiple livekit servers and multiple egress servers, deployed in different regions. The ClusterID in each livekit server config needs to match a ClusterID in an egress config

Thanks <@U01UB0LHTTQ> that makes sense, I rushed trying to make sense of it and later realized my second node was able to handle egress requests too :))

I have added the firewall rules,m but to no avail

In one of my implementation I need to whitelist all the domains, so I was curious to know if there was a way to know all the subdomains that the server needs?

hey <@U032TUAL4KT> - that url is not new, but I I think we need to update our <https://docs.livekit.io/cloud/firewall/|firewall doc>

Thanks for the info <@U02BBTRGF98>! My bad. It seems like it was the first time that had seen these. I'll update my app based on those domains. Once again, thank you

in addition to `sfo3.production`, you may want to also open up:
• gapnortheasta.production
• gapsoutha.production
• geucentrala.production
• gsaeasta.production
• guseasta.production

Its rather that i need to whitelist all the domains which my app is allowed to reach. Hence why I need that list. Once again thanks

gotcha! you should also whitelist all of our TURN domains too then. that’s the same as above, but replace “production” with “turn”

This can either be a CPU issue (can't record audio fast enough) or a connection quality issue. Are you using cloud or open source?

We are facing the same issue with cloud egress, Livekit client sdk

I've just added some debugging and profiling tools to egress, will look into it this week to see if the issues are coming from egress, or the publisher or sfu

Are there any egress events to find recording issues and audio track loss at client side

can you guys share an example of what you mean by audio being jittery? This is a Web Egress? what are you recording?

Recorded video in egress, audio is not in sync with video/ audio is unavailable after some duration in egress video recording

<@U05E0KG6YQY> do you mind DMing me an Egress ID when this had occurred? we can take a look on our side.

Hey <@U01KZ61GV5Z> sharing the sample in DM

logs.log

Still working through it, logs are attached.

Refreshing client token after claims change appears to race. The version on ParticipantInfo is incrementing, but not sure if LKServer can see success in dispatching the event to client.

We do have selective subscription/publication being managed client-side, but it still happens when turned off.

Pls thread or edit you original message :pray::skin-tone-4:  

logs.txt

In the first log, it is due to packet loss
```pe: 559, pl: 81,```
that is 81 packets lost out of expected 559, that is ~15% loss. That is high loss.

The second one is due to high jitter.

Which version of server are you on? Please update to the latest. We have removed jitter as we found it to be too volatile and gives false alarm a bunch. Similarly for RTT also.

But, high jitter values do indicate a network issue. Looks like there are periods of significant loss too. So, overall not a very stable network which could explain the observed quality.

Latest version  v1.4.3.

My helm config:
image:
  tag: v1.4.3

oh my bad, the second log also has significant loss
```pe: 305, pl: 109```
That is almost 33% loss. So, loss is the reason for the quality drops.

The quality is not bad, but there is packet loss, right?

Yeah, that looks like good chunks of packet loss. Losses need to be repaired, either through NACK from receiver indicating which packets were lost and retransmission by sender OR receiver asking for a key frame as it cannot decode the packets it has (broken sequence due to losses). In either case, it is a disruption (in this case, looks like it has to get key frames, guessing).

Is the problem the publisher's internet? Why is there packet loss?

Yeah it is on upstream from publisher to the server. So it could be their Internet

The same publisher is using multiple platforms and these platforms are using agora rtc and tencent rtc.
There is no freezing on other platforms, so it makes sense that packet loss happens when using livekit.

We have also tested 3 different applications using livekit, tencent and agora service, only Livekit has packet loss.

Tough to say why that could be without the full deployment picture. Maybe the connectivity from the publisher’s location to where you have deployed LK is not great?

There is a 50ms ping from the publisher's location to the server.
My server's speed test: <https://www.speedtest.net/result/c/ee241cbb-9d79-4260-a382-7b86518ddca6>

publisher's speed test:

<@U041KCGQCGZ> we really can't say as you are self hosting. hosting providers differ drastically in terms of udp delivery even tho they could maintain a low ping. We have seen this happen with discount hosting providers.

if you are comparing to other cloud vendors, it would be better to compare LK Cloud.

related. these speeds look rather low for a server.  we typically see 1Gbps+ on a server.

Server has 10Gbps network : <https://www.speedtest.net/result/c/ee241cbb-9d79-4260-a382-7b86518ddca6>

<@U041KCGQCGZ> do you see the same issues when using <https://cloud.livekit.io|LiveKit Cloud>? You might want to switch for testing purposes. We have a free plan with 50GB of bandwidth included - no credit card required.
Switching should be as easy as changing the server url, api key and secret. If the problem persists, we have more debugging and analysis options to help you quickly.

yes, we will test it with Livekit Cloud, I will report the result.

<@U03RM7G8Q6A> <@U01KZ61GV5Z> <@U02CM28PB4Y> We tested. Video freezes are also available in Livekit Cloud. What information is required for your debugging?

<@U041KCGQCGZ> can you send us a link to a session that exhibited issues for you?

<@U01KZ61GV5Z> <https://cloud.livekit.io/projects/p_47iqx46bhs7/sessions/RM_HLgVuZthVxb5>

There is significant packet loss from the publisher side - <https://cloud.livekit.io/projects/p_47iqx46bhs7/sessions/RM_HLgVuZthVxb5/participants/PA_fG3SMFQxekTL>.

Looks like the path from publisher -&gt; data centre in Frankfurt is experiencing lots of loss.

Currently we do not have a data center in your region. However, we are considering bringing up a new region in Tel Aviv.. that should improve the experience significantly for you then.

I'm curious, where are you hosting right now? when you ran the test against self-hosted.

<@U01KZ61GV5Z> Our servers are hosted in Falkenstein/Germany.

Publisher's internet speed is very good, ping 50 ms. Except for this person, this problem is not observed in 20 different broadcasters in approximate locations.

right, I trust that the speed test is looking decent. However, this is a bit more complicated than that. Speed test is pointing the user to the closest edge server (often hosted by the same ISP). The goal of a speed test is to measure their connection from the ISP to their computer.

When you are hosting a server, you'd have to look at how the traffic flows from your server to the user's ISP. And this is really a complicated web of inter-connectivity and peering agreements that most of us don't tend to think about. Based on the session you had shared earlier, it doesn't look like the connection from Germany to the user in Turkey was very good.

Could packet losses be caused by the Flutter library? Is packet loss possible due to the code structure that some phone models do not support?

Let's not forget that the user has the same packet loss rate with both wifi and cellular data. Does this make any sense to us?

it's unlikely. all of our native SDKs run libwebrtc, the same code as Chrome when it comes to communications.

We would like to try the livekit cloud again with the servers you will set up in Tel Aviv and look at the packet loss problem.

how do you `remove track in browser`? what livekit API does that translate to?

```await this.room.localParticipant.unpublishTrack(track)```

it works for other users connected in browser
only happens in server

there does appear to be an issue here. taking a look

this has been fixed in 1.0.13. thanks for the report! we had a regression at some point and the bug had existed for awhile now

Hey <@U04S9JQDA8Y> can you describe what the issue is?

I'm currently using egress with docker up and using a vm with 8cpu 32GB specs. Currently, after completing the socket connection to the room, when I request egress to start recording the room, egress is terminated and the recording is not recorded for more than a certain number of seconds. <@U01KZ61GV5Z>

what kind of egress are you performing? it'd be best to look at Egress logs for clues

```2023-06-27T08:59:33.660812458Z 2023-06-27T08:59:33.660Z INFO    egress  source/web.go:240       chrome: START_RECORDING {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G"}
2023-06-27T08:59:35.983356953Z 2023-06-27T08:59:35.983Z INFO    egress  pipeline/watch.go:152   pipeline playing        {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G"}
2023-06-27T08:59:35.983395928Z 2023-06-27T08:59:35.983Z INFO    egress  service/handler.go:214  egress updated  {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G", "egressID": "EG_M6aU8bDBhe6G", "request_type": "room_composite", "output_type": "file", "status": "EGRESS_ACTIVE"}
2023-06-27T08:59:48.670726892Z 2023-06-27T08:59:48.670Z INFO    egress  source/web.go:250       chrome: END_RECORDING   {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G"}
2023-06-27T08:59:48.670749202Z 2023-06-27T08:59:48.670Z INFO    egress  service/handler.go:214  egress updated  {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G", "egressID": "EG_M6aU8bDBhe6G", "request_type": "room_composite", "output_type": "file", "status": "EGRESS_ENDING"}
2023-06-27T08:59:48.671413679Z 2023-06-27T08:59:48.671Z INFO    egress  pipeline/pipeline.go:382        sending EOS to pipeline {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G"}
2023-06-27T08:59:48.685403444Z 2023-06-27T08:59:48.685Z INFO    egress  pipeline/watch.go:68    EOS received, stopping pipeline {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G"}
2023-06-27T08:59:48.691829418Z 2023-06-27T08:59:48.691Z INFO    egress  pipeline/pipeline.go:498        pipeline stopped        {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G"}
2023-06-27T08:59:48.936383283Z 2023-06-27T08:59:48.936Z DEBUG   egress  sink/file.go:59 removing temporary directory    {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G", "path": "/tmp/lkegress/EG_M6aU8bDBhe6G/"}
2023-06-27T08:59:48.936458405Z 2023-06-27T08:59:48.936Z INFO    egress  service/handler.go:208  egress completed        {"nodeID": "NE_mqpkJ9RAHmEL", "handlerID": "EGH_cAGVmo9siBh3", "clusterID": "", "egressID": "EG_M6aU8bDBhe6G", "egressID": "EG_M6aU8bDBhe6G", "request_type": "room_composite", "output_type": "file"}```
here is my egress logs, and i use roomComposite recording, and also I dont request egress ended after connection finished. <@U01KZ61GV5Z>

Changed the port range and relay port range and it worked! As expected, firewall stuff.

it's not very resource intensive at all. we have some guidelines <https://docs.livekit.io/oss/benchmark/#measuring-performance|here> along with the tooling you'll need to measure performance for your specific use case.

Ah okay, sweet! I'll check it out. Thank you so much!

<@U05603TMKCZ> When you use flutter to share the screen, can you use another livekit client to see the 1:1 picture? I guess that egress laid out a vertical android video on a horizontal recording canvas.

Or if you can upload the recorded video clip to better analyze the issue.

Hmmm so what we do is one publishes shared screen and microphone and the others only publish microphone. Is there anyway to change the recording canvas?

is this our default recording templates? what is the Egress request here? i.e. the full `StartXXXEgress` API call and all of its parameters (feel free to redact the keys/etc)

You can see this <https://mathtutor-wim-staging.s3.amazonaws.com/video/19c32b11-fa24-4e23-a5bd-38f7ff5cfd61.mp4|video>. Is this the template problem? I'll tell our backend developer then.

Hmm, the Portrait video drawn on the Landscape canvas, maybe add a static image background of another color? Or set up a proprietary Portrait canvas template for screen recording?

I think you are probably using the default output settings.. which records a landscape video.

please see <https://docs.livekit.io/server/egress/#encodingoptionspreset|docs here> on passing in the right encoding preset. you'd want one of these that start with `PORTRAIT_`

I think that's what the problem was. Thank you very very much for your help. Your solution is the best and so is your support!

we don’t expose the attached video views from the video track itself, that’s something you’d have to keep track of on your end

The addrenderer function expects a videosink object, is videosink a standard webrtc class or is it some wrapper implemented in the livekit sdk?

VideoSink is a standard webrtc interface, Surface/TextureViewRenderer both implement VideoSink

Hey <@U05D8952VPC>, there is a limit on “Total concurrent participants” of 100

You can check if you hit it in the Dashboard &gt; Settings &gt; Project &gt; Connection limits

limitation.png

Hi <@U03RM7G8Q6A>,

Thanks for your response. Our project hasn’t hit any limit yet

I found as soon as our meeting duration hit 5 mins, livekit websocket get disconnected and closed the meeting

This is not expected. Can you DM me your project ID? I will forward it for internal investigation.

Can you check if you can reproduce the same behavior with Meet? <https://meet.livekit.io/?tab=custom>

We tried it with <http://meet.livekit.io|meet.livekit.io>. It doesn’t cause similar issue.

<@U05D8952VPC> is there a reason why you set `recorder: true` when generating the token?

We are simply just want to indicate the meeting is being recorded. 

That would be great :+1:
<@U022KB72NFJ> suspects:
&gt; Yes my guess is that if recorder is true, the room will autoclose because the SFU thinks there is no participant inside it

When the meeting is being recorded with LiveKit Egress, `Room.isRecording` will automatically flip to true.

Our recording stack is built by having a new participant join the session with  `recorder: true` grant.

that grant should not be used by normal participants

Thanks team for your prompt support. Yes, setting `recorder: true` cause the issue of meeting disconnect after 5 mins.

I'm also rather new to LiveKit, but I've seen their spatial audio example, that may help you with the music streaming bit <https://github.com/livekit-examples/spatial-audio>

For the queuing system, you can use <https://docs.livekit.io/client/data-messages/#participant-metadata|participant metadata> to handle this. When a participant enters the queue, you could include a field with the current timestamp. That should allow you to sort the queue on the client side.

On synchronization, yes, it'd be better to ensure that music and user's singing come from the same machine. That is the best way to ensure there are no unexpected latency between different sources.

Thank you very much for your responses and the time you have given.

Hey <@U05DULWKW9M>,

For moderation, I would recommend checking out the server APIs.
• <https://docs.livekit.io/server/room-management/#mutepublishedtrack|MutePublishedTrack> - mutes another participant
• <https://docs.livekit.io/server/room-management/#updateparticipant|UpdateParticipant> - set the other user's permission object to include `canPublish`, `canPublishSources`, etc
The typical flow we recommend is having your server be in control of who has permissions to perform these operations. And then calling our server APIs to make those changes.

We are also in the process of building out an example app to demonstrate participatory livestreaming, where you could have multiple hosts streaming to a large audience. (i.e. Clubhouse with video)

Thank you for taking the time help me out!

I've seen the server API and I intend on having the server control the permissions.

I'm just not sure how I can use the hooks and components provided to build a custom UI close to the meet example where in each custom `ParticipantTile` it has some buttons like the `ControlBar` that the admin can control the permissions of that individual participant.

I'm sure that's doable, I just can't seem to find the right components or hooks to customize the experience.

currently, we don't have built-in components that would do exactly what you want (giving admins control over the situation). However, this should be fairly easy to build yourself. You can start with `ParticipantTile`, but then override the children to provide custom rendering of your own components. Currently, the `ControlBar` is designed to work with a local participant's tracks, however, you could override the buttons so that they are making server API calls to your backend server.

please only post to one channel, thanks! answered in <#C025KM0S1CK|dev>

Hey there, can you please drop a quick comment on how `livekit_participant_join_total` metric in Grafana should be read: `rtc_connected`, `rtc_init`, `signal_connected` - what are these. And I cannot see any TURN-related metrics, thanks !

this looks like maybe the viewer isn't subscribing to the right thing ?

<@U05CKJW1W67> also posted (pls thread related messages):

&gt; i am seeing the samne behavior using the example demo hosted by livekit.

hi Jonas, could you share more details re: your livestreaming setup? are you using OBS?

also from the viewer POV, they may need to click on the video player to enable audio (it's disabled by default on browsers like Chrome)

<@U03GQN6KXCG> i used OBS to stream using the endpoint displayed and steramkey, i used h264 for codec.
on the viewer side i did click on the video to unmute it.

gotcha. let me see if I can reproduce it on my end!

<@U03GQN6KXCG> here is the link to a channel i am currently streaming to : <https://livestream-demo.livekit.io/channel/jonastest>

k I also see an empty video on my end. I'll take a deeper look today - the repo hasn't been updated in months and there might have been a breaking change since then :pray:

Hi <@U05CKJW1W67> :wave: we've identified and fixed the bug that broke viewing livestreams. you can pull the latest changes from the demo repo!