Troubleshooting “No available render server, Please try again later.”

This error usually means that LarkXR cannot assign an available render node to the session request. The most common causes are offline render nodes, exhausted node resources, whitelist issues, or render nodes being occupied by remote desktop usage.

## Symptom

The web client displays the following message:

> No available render server, Please try again later.

## Cause 1: No Online Render Nodes

Log in to the management backend and open **Data Center > Resource Monitoring**. If the number of online nodes is `0`, no render server is currently available.

![Resource monitoring example](https://docs.paraverse.cc/media/202504/5815b613139b4b16bf3143e5f73c9c7e5814.png)

### Solution

1. Check whether the rendering service is running.
2. Confirm that the render node IP address has been added to the LarkXR whitelist.
3. Check the render service configuration and verify that the Nacos registry IP address in the Render Manager configuration is correct.

Reference links:
- [How to start the rendering service](https://www.pingxingyun.com/contribute/detail?topicId=10106&categoryId=3)
- [How to configure the IP whitelist](https://www.pingxingyun.com/contribute/detail?topicId=10109&categoryId=3)
- [Rendering service configuration reference](https://www.pingxingyun.com/contribute/detail?topicId=10106&categoryId=3)

## Cause 2: Node Resource Usage Is Too High

If GPU, CPU, or memory usage on a render node reaches the configured threshold, LarkXR will stop assigning new sessions to that node. The current page states that the default behavior is to stop further allocation when any of these resources reaches 90%. [page:10]

### Solution

Adjust the resource threshold settings in the management backend if the current limits are too strict for the deployment scenario.

Reference link:
- [How to configure resource thresholds](https://www.pingxingyun.com/contribute/detail?topicId=10108&categoryId=3)

## Cause 3: Render Resources Are Occupied by Remote Desktop

If someone is using the LarkXR backend remote desktop feature, an exclusive application may report insufficient rendering resources because the render node is already occupied.

### Solution

1. Log in to the management backend.
2. Open the **Process Management** page.
3. Stop the process currently using remote desktop resources.
4. Try launching the session again.

## Notes

- Start by checking **Data Center > Resource Monitoring**, because this is the fastest way to confirm whether the problem is caused by offline nodes or unavailable capacity.
- If online nodes exist but sessions still cannot be assigned, review both the whitelist configuration and the current resource usage before checking application-level issues.
- If the problem affects only exclusive applications, also verify whether the render node is being held by another remote desktop session.