feat: filter NodeReadiness alerts on uncordoned status

[TheRealNoob]

Hello,

First time contributing to this repo and writing jsonnet so I welcome any feedback on the syntax that I've used here.

In my personal experience at #dayjob we get many false alerts on nodes which are undergoing maintenance. I believe filtering out cordoned nodes from these alerts is something many people would benefit from. Thoughts?

Thank you

[skl]

Thanks, this looks like a reasonable change.

Please update this test:

kubernetes-mixin/tests.yaml

Lines 571 to 586 in af5e898

	- interval: 1m
	input_series:
	- series: 'kube_node_status_condition{condition="Ready",endpoint="https-main",cluster="kubernetes",instance="10.0.2.15:10250",job="kube-state-metrics",namespace="monitoring",node="minikube",pod="kube-state-metrics-b894d84cc-d6htw",service="kube-state-metrics",status="true"}'
	values: '1 0 1 0 1 0 0 0 1 0 1 0 0 0 1 0 1 0 0 1'
	alert_rule_test:
	- eval_time: 18m
	alertname: KubeNodeReadinessFlapping
	exp_alerts:
	- exp_labels:
	cluster: kubernetes
	node: minikube
	severity: warning
	exp_annotations:
	summary: "Node readiness status is flapping."
	description: 'The readiness status of node minikube has changed 9 times in the last 15 minutes.'
	runbook_url: 'https://github.com/kubernetes-monitoring/kubernetes-mixin/tree/master/runbook.md#alert-name-kubenodereadinessflapping'

Extra credit if you can create a test for KubeNodeNotReady! 😄

[TheRealNoob]

I think I did this right? I'd really appreciate a second set of eyes as its my first time using the native prometheus unit tests. Appreciate the advice you gave me on Slack, it was super helpful!

[skl]

lgtm, I like that you used two simple node series where only one node is expected to have an alert firing.

[skl]

Shorthand syntax suggestion that you might like to know, for next time:

Suggested change

	values: '1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1'
	values: '1x19'

docs

Which would translate to:

shorthand for '1+0x19', series starts at 1, then 19 further samples incrementing by 0.