Ayoub Mrini

I conducted some benchmarking on an "idle" cluster, but the results were inconclusive. Aside from a significant and continuous increase in CPU usage, I did not observe any notable decrease...

> In a large k8s cluster, we want to monitor pods running on a subset of nodes. Relabeling is costy. Most pods are dropped by relabeling and the process causes...

Unfortunately we're still seeing this on `main` ``` === RUN TestAsyncRuleEvaluation/asynchronous_evaluation_of_all_independent_rules,_insufficient_concurrency manager_test.go:1963: Error Trace: D:/a/prometheus/prometheus/rules/manager_test.go:1963 Error: "0.8259941" is not less than "0.09" Test: TestAsyncRuleEvaluation/asynchronous_evaluation_of_all_independent_rules,_insufficient_concurrency === RUN TestAsyncRuleEvaluation/asynchronous_evaluation_of_all_independent_rules,_sufficient_concurrency --- FAIL: TestAsyncRuleEvaluation...

https://github.com/prometheus/prometheus/pull/14300 was merged for this.

I'd like to get the approval back @bboreham.

With: ``` scrape_configs: - job_name: endpoints kubernetes_sd_configs: - role: endpoints kubeconfig_file: PATH/TO/CONFIG ``` Where the kubeconfig contains an unresolvable apiserver. Prometheus [8655fe5](https://github.com/prometheus/prometheus/commit/8655fe5401c79e1fc176ca8b4644db04462dc9f5) does continually forward k8s client logs about that...

https://github.com/prometheus/prometheus/issues/11574 is asking about metrics. I'll take a look at that.

We're really interested in this feature. Can I help somehow?

Ok, I'll give it a try, thanks.

AFAIK (for K8s and also for OpenShift) using `--swap-deployment`, Telepresence proxy deployment will use the old deployment's `spec.containers[].resources`; if pods of postgres-gis are already running in the cluster there is...