I am getting
forceDisconnectByServer and ghost disconnections for my implementation of OV.e[36mopenvidu-
The max number of users I was able to get inside a single session was - 11 and at that point, I started to get these disconnections.
server_1 |e[0m [INFO] 2021-03-16 15:08:53,629 [http-nio-0.0.0.0-5443-exec-3] io.openvidu.server.kurento.core.KurentoSessionManager - Request [LEAVE_ROOM] for participant ipc_IPCAM_rtsp_Y2LO_10_7_24_11_554_profile2 of session 5-1615887420306 with reason forceDisconnectByServer
Extra info:
I am hosting all assets using multiple docker containers inside a single 16core instance.
I am streaming my participant’s canvas to each other.
And often use IP cam from time to time.
OV version is 2.16
@pabloFuente tagging you for visibility - this issue is in our production and would love to have your insights here.
forceDisconnectByServer reason appears only when your backend application calls method DELETE Connection: OpenVidu Docs
Okay then it might be that I am reading the logs wrong. But we are noticing high frequency of disconnections as pointed in the beginning. From your point of perspective where we should start investing and what could be the possible reasons behind it ?
As pointed out this is a 16 core server with only one session in it.
Do you think lack of cpu power on client side could be the issue ? (These are all modern tab and Android devices)
How have you deployed OpenVidu? What does this exactly mean?
I am hosting all assets using multiple docker containers inside a single 16core instance.
Did you deployed OpenVidu as stated in the documentation? : OpenVidu Docs
Sorry for the confusion
I meant was our application server and OV instance are running inside the same box. In their own docker instances.
So this is your OpenVidu deployment: OpenVidu Docs