New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Flaws in the definition of apiserver metrics #690

Open

allenhsu opened this issue Nov 3, 2021 · 0 comments

allenhsu commented Nov 3, 2021 •

edited

Loading

Related issue: prometheus-operator/kube-prometheus#1480

https://github.com/kubernetes-monitoring/kubernetes-mixin/blob/master/rules/kube_apiserver.libsonnet#L50

Before any 5xx error happens, this metric will return no data.

Similar to the or vector(0) in the definition for availability metrics.

sum by (%(clusterLabel)s) (rate(apiserver_request_total{%(kubeApiserverSelector)s,%(kubeApiserverReadSelector)s,code=~"5.."}[%(window)s]))

should be:

sum by (%(clusterLabel)s) (rate(apiserver_request_total{%(kubeApiserverSelector)s,%(kubeApiserverReadSelector)s,code=~"5.."}[%(window)s]) or vector(0))

The text was updated successfully, but these errors were encountered:

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment