[Bug]: NIC will do reload for endpointslice changes even you used the plus version

[wd]

Version

3.7.2

What Kubernetes platforms are you running on?

Rancher

Steps to reproduce

NIC will try to do a batch reload when the queue size is greater than 1: here. This is great because it can reduce the reload times.

NIC will use the API to update endpoints for the plus version. However, the batch reload will also be enabled when multiple Endpointslices changes occur simultaneously for the plus version. This will happen frequently when you have a busy cluster.

This is the CPU idle graph of one of our Ingress nodes. Because of this issue, the CPU usage is not stable.

[github-actions]

Hi @wd thanks for reporting!

Be sure to check out the docs and the Contributing Guidelines while you wait for a human to take a look at this 🙂

Cheers!