code202

Reason One (Primary):：

When using proxy mode, the stream is pushed through the WebSocket address ws://{{management node IP}}/{{render node IP}}/10002 to maintain a long connection. Network disconnections (initially unable to connect) or network instability (encountering error 202 during use) can lead to error 202.

Troubleshooting and Solutions for Network Disconnections:：

1. Intranet Environment：

   • Check Firewall Settings: Ensure that the firewall is turned off or properly configured to allow network traffic.
   • Verify Local IP Address Configuration on Render Node: Confirm that the local IP address configured on the render node matches the actual network environment.

   Refer to the following documentation：

   • Standalone Intranet(windows)
   • Cluster Intranet(windows)

2. Extranet Environment：

   • Check Preferred Public IP Settings: Ensure that the preferred public IP is correctly set to the external IP address to allow normal external access.

   Refer to the following documentation：

   • Standalone Extranet(windows)
   • Cluster Extranet(windows)

By following the troubleshooting directions and solutions above, you can systematically address and resolve network disconnection issues for both intranet and extranet environments to ensure normal system network connectivity.

###Solution for Network Instability：

To address network instability in private deployment scenarios, you can implement front-end proactive reconnection to maintain a stable connection. The specific steps are as follows:

1. ** Enable Application Reconnection in Management Console：**

   • In the management console settings, you can configure the application reconnection feature through the path: Application Management >> Edit >> General Advanced Settings. As shown in the figure below: select the delay stream delay close, default is 20s.。

2. Modify Configuration File Larkxr-center\application.properties :

   • Adjust the parameters in the file to extend the task heartbeat duration to prevent tasks from being released when the server disconnects. Restart the Admin service after making changes.

# Default 12000 milliseconds to take a node offline without a heartbeat
# When setting task delay close, this can be changed to the duration configured in the backend, for example, 120000 (2 minutes)
pxy.clear-job.server-no-heartbeat-duration=120000
# Default 20000 milliseconds to close TaskInfo without a heartbeat
# When setting task delay close, this can be changed to the duration configured in the backend, for example, 120000 (2 minutes)
pxy.clear-job.taskinfo-no-heartbeat-duration=120000

2. Web Front-End Modifications：
   Iframe:
   【Attachment】datachannel.js
   WebSDK：
   【Attachment】websdk_resolve202.zip

Reason Two：

The rendering package was mistakenly terminated, resulting in its absence. The complete structure of the program package is shown below:

• Solution：

Re-download the program package and replace the contents of the RenderServer directory.

Reason Three：

When using a GeForce graphics card, the number of hardware encoding channels is limited by Nvidia itself. Connecting to remote tools like Sunflower will also occupy one encoding channel.

• Solution：

  • Consider using Quadro or Tesla graphics cards, or opt for the provided all-in-one hardware and software solution to increase the number of encoding channels.
  • Quadro and Tesla graphics cards typically offer more hardware encoding channels, making them suitable for high-concurrency encoding needs.
  • Consider using an all-in-one hardware and software solution for more efficient encoding performance and additional concurrent channels.