The dashboard of the large cluster gets stuck on the embedded_back_pressures call.

[shanicky]

Is your feature request related to a problem? Please describe.

In a larger cluster, opening the dashboard on port 5691 often results in requests getting stuck at a global embedded_back_pressures interface, which frequently prevents the page from loading.

Consequently, it becomes impossible to access the fragment_graph interface, which is often the only way to observe the internal structure of the fragment.

Describe the solution you'd like

No response

Describe alternatives you've considered

No response

Additional context

No response

[BugenZhao]

Consequently, it becomes impossible to access the fragment_graph interface

Interesting. Theoretically, retrieving the back-pressure rate should be done asynchronously without affecting the loading and rendering of the fragment graph. Could you please provide the steps to reproduce the issue?