Pranshu Srivastava

Right, boolean flags surely are a possible solution. However, my inclination towards replacing them with feature gates was motivated by a couple of observations. - The feature-gates model implemented here...

That being said, if you think we are better off without these, PLMK and I'll close this.

@SuperQ @discordianfish I'd be happy to add this to the agenda and pitch it in the call next-to-next week if that sounds good (or have an async discussion here if...

I believe this should establish a robust and streamlined way of gating features here, re-pinging @SuperQ and @discordianfish for a look. 🙂

This failed to gain traction. Closing.

@discordianfish Please correct me if I'm wrong but I believe even after _fixing_ https://github.com/prometheus/procfs/pull/178#discussion_r1012799750, i.e., not opening a file that's not read, we'd still have to figure out a way...

Can we include https://github.com/kubernetes/kube-state-metrics/pull/2418 as well?

Seconding Damien, I'm not for panicking the binary since we expect a [reload](https://github.com/kubernetes/kube-state-metrics/blob/f8aa7d9bb9d8e29876e19f4859391a54a7e61d63/internal/wrapper.go#L61-L62) on configuration change, and then a log (or metric) indicating an issue with the given configuration. Crashing...

Exactly what Damien said. Additionally, PTAL at the docs, https://github.com/kubernetes/kube-state-metrics/pull/1851/files#diff-1e83b132955ba4b76949317660faf1fa222f0ee999d4fb3b4fb9dd318b7ac2beR521.

Native resources are supported in-tree only, by design, and thus cannot leak into the CRS (_**Custom Resource**_ State) machinery. If you believe something needs to be added for them, please...