Marco Pracucci

> This is equivalent in TSDB but not in PromQL, so the optimization could be done in tsdb. I think it would be great. I think we tried to do...

> It seems that we should use `!=""` rather use `.*` or `.+` in the most situations now. If you're currently using `app=~".*"` it also matches metrics without the `app`...

I'm really wondering how many Prometheus users are aware that `.*` or `.+` doesn't match if the label value contains a newline. I believe the expected behaviour is that it...

@roidelapluie I think this PR introduced an issue in JSON marshalling. Could you take a look at https://github.com/prometheus/common/pull/651 please?

This looks like something to build on top of Cortex (eg. automation to generate alertmanager config based on CRD) and not something to do within Cortex. I think supporting Kubernetes...

We have the next Cortex community call scheduled for 1st April. If it's not too late, I would suggest you to join it and discuss it in that occasion, otherwise...

> Thanks, are these the relevant details for the Cortext community? @mjallday Yes!

From the [original comment](https://github.com/grafana/mimir/pull/8886#issuecomment-2265802337), you proposed 3 options: > * Always use the mapped metadata cache for user issued `Metadata` requests > * Introduce a new API, `RequestCachedMetadata(req *kmsg.MetadataRequest, expiry...

A couple of requests, please: 1. Could you provide an example that shows the issue you're fixing, please? 2. Can you check the CI linter failure, please?

Thanks Bartek for the feedback! > Still I think we could implement the design proposed by @pracucci and then allow swapping hash function with some special hash that ignores replica...