usage-metrics-collector icon indicating copy to clipboard operation
usage-metrics-collector copied to clipboard

High fidelity and scalable capacity and usage metrics for Kubernetes clusters

Results 8 usage-metrics-collector issues
Sort by recently updated
recently updated
newest added

Per https://testgrid.k8s.io/sig-instrumentation-usage-metrics-collector#periodic-test&width=20 sigs.k8s.io/usage-metrics-collector/cmd/metrics-node-sampler/integration.TestMetricsNodeSamplerGRPC/basic test is frequently flaking. Example failure: https://prow.k8s.io/view/gs/kubernetes-jenkins/logs/ci-usage-metrics-collector-test/1765441415891914752 Diff: --- Expected +++ Actual @@ -169,2 +169,3 @@ "cpuCoresNanoSec": [ + "1000", "1000" @@ -172,2 +173,3 @@ "memoryBytes": [...

help wanted
kind/flake
triage/accepted

From https://testgrid.k8s.io/sig-instrumentation-usage-metrics-collector#periodic-test&width=20 sigs.k8s.io/usage-metrics-collector/cmd/metrics-node-sampler/integration.TestMetricsNodeSamplerRestEndpoint/basic is pretty regularly flaking. Example failure: https://prow.k8s.io/view/gs/kubernetes-jenkins/logs/ci-usage-metrics-collector-test/1765199318811152384 Diff: --- Expected +++ Actual @@ -158,2 +158,3 @@ "cpuCoresNanoSec": [ + "2000", "2000" @@ -161,2 +162,3 @@ "memoryBytes": [...

help wanted
kind/flake
triage/accepted

Tried giving this a run-through but the readme seems incomplete, if I follow it verbatim, I end up SET_THIS:v0.6.0 everywhere and things obviously unable to run.

triage/accepted

Closes #138. #### Goals: - don't get rid of cgroupv1 support - add support for cgroupv2 for all existing metrics pulled out of cgroups - update both the direct cgroup...

approved
cncf-cla: yes
do-not-merge/work-in-progress
size/XL
needs-triage

Currently, UMC only supports collection of metrics with cgroupv1. cgroupv2 has been [GA in Kubernetes since 1.25](https://kubernetes.io/blog/2022/08/31/cgroupv2-ga-1-25/), so it would be nice to add support. /kind feature

kind/feature
triage/accepted

While using `testutil.TestDir` in a project to run unit tests against a local kube-apiserver, I noticed that I was building up orphaned etcd and kube-apiserver processes on each test run....

size/S
ok-to-test
cncf-cla: yes
triage/accepted

I found an error during the compilation of the Docker image, following the steps in the README documentation. Is this error because the go version is not forward compatible? go...

help wanted
kind/support
triage/accepted

size/S
cncf-cla: yes
do-not-merge/work-in-progress
needs-triage
needs-ok-to-test