Some stream will occur: ICE connection state changed to closed. Can someone tell me how to solve this?

[1421855993]

This maybe occur in every client.

[GaoMingKai]

Please give me the detailed steps leading to ICE connection state changed to closed

[hubiwibe]

I have a same problem.

The detailed step leading to ICE Connect state changed closed is We are testing with javascript conference basic example. The testing environment is Server : Release 4.3 (CS_WebRTC_Conference_Server_MCU.v4.3.Ubuntu.tgz) (Ubuntu 18.04) Client : Chrome (version 78.0.3904.87) in Ubuntu 18.04, Chrome (most recent version) in Windows, Chrome in Android, Safari in iOS

Everytime we connect to the sample page (https://192.168.0.45:3004) with MCU option or (https://192.168.0.45:3004/?forward=true) with SFU option, after 2 or 3 minutes, "ICE connection state changed to disconnected" and "ICE connection state changed to closed" happens. (192.168.0.45 is our server ip address) (See the attached image for reference)

[cisco66666]

@Yungi-Jung @Yungi-Jung I saw the steps, very detailed, but some details would like to ask，Are you all connected to mcu at the same time in the above four clients, and pub and sub are a mix and a forward stream? The last question will occur for 2-3 minutes, right?

[hubiwibe]

@cisco66666 @cisco66666 Thank your for your comment.

Yes, four clients are connected to the mcu server at the same time. If I understand your question correctly, First time the four clients pub and sub with forward stream. Second time the four clients pub and sub with mix stream.

We didn't pub and sub with forward and mix at the same time.

The problem occur within 2 minutes after the four clients connected to the server. Sometimes it happens immediately after the four client connected to the server.

[cisco66666]

I tried 4 clients (the Android side, the chrome network has a problem, so use windows chrome instead) to connect mcu, and pub a stream, sub all other forward streams, wait 10-20 minutes, no duplicates are idle The problem. Here, the Android network debugging is good, and then re-test, you also try to use Android, there will be no problem？

[hubiwibe]

@cisco66666 Thank you for your answer.

After reinstalling os, I confirmed that it works.

[hubiwibe]

@cisco66666 Hello cisco66666.

I did a sfu connection test with 8 devices in a Wi-Fi environment without modifying the source code.

• devices : iPhone 3ea, Android phone 3ea, laptop 2ea
• wifi model : IpTime A2008
• sfu : management console > removed views, connection to 3004/?forward=true

Discoonection occurs intermittently from the seventh connection.

1. Is this normal?
2. Can I know the maximum number of concurrent users?

[linewei]

i have the same problem.

[linewei]

Hi, Yungi-Jung @Yungi-Jung Can you share your solution? I met the same problem, in almost the same environment.

[hubiwibe]

Hi, Yungi-Jung @Yungi-Jung Can you share your solution? I met the same problem, in almost the same environment.

Hi, linewei @linewei It was solved after OS and OWT server reinstallation. However, it can happen if the network status is not good. For example, this occurred when eight mobile devices were connected to one room.

[Chandramouli-P]

Hello Team,

We are also facing the same issues and posted in Intel CS for WebRTC forums also. We already tried couple of times, as Yungi-Jung said. But, I strongly believe that this is not the solution. Hence, request you to help us to resolve the issue.

Thank you.

Best Regards, Chandramouli.