chore: Update the build_info metric label naming and tests (#5275)

charts/karpenter/README.md

@@ -34,7 +34,7 @@ helm upgrade --install --namespace karpenter --create-namespace \
 | controller.envFrom | list | `[]` |  |
 | controller.extraVolumeMounts | list | `[]` | Additional volumeMounts for the controller pod. |
 | controller.healthProbe.port | int | `8081` | The container port to use for http health probe. |
-| controller.image.digest | string | `"sha256:afa0d0fd5ac375859dc3d239ec992f197cdf01f6c8e3413e3845a43c2434621e"` | SHA256 digest of the controller image. |
+| controller.image.digest | string | `"sha256:5e5f59f74d86ff7f13d7d80b89afff8c661cb4e3265f2fdda95b76dd9c838cc1"` | SHA256 digest of the controller image. |
 | controller.image.repository | string | `"public.ecr.aws/karpenter/controller"` | Repository path to the controller image. |
 | controller.image.tag | string | `"v0.33.0"` | Tag of the controller image. |
 | controller.metrics.port | int | `8000` | The container port to use for metrics. |

go.mod

@@ -15,6 +15,7 @@ require (
 	github.com/patrickmn/go-cache v2.1.0+incompatible
 	github.com/pelletier/go-toml/v2 v2.1.0
 	github.com/prometheus/client_golang v1.17.0
+	github.com/prometheus/client_model v0.5.0
 	github.com/samber/lo v1.39.0
 	go.uber.org/multierr v1.11.0
 	go.uber.org/zap v1.26.0
@@ -79,7 +80,6 @@ require (
 	github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 // indirect
 	github.com/olekukonko/tablewriter v0.0.5 // indirect
 	github.com/pkg/errors v0.9.1 // indirect
-	github.com/prometheus/client_model v0.5.0 // indirect
 	github.com/prometheus/common v0.44.0 // indirect
 	github.com/prometheus/procfs v0.12.0 // indirect
 	github.com/prometheus/statsd_exporter v0.24.0 // indirect

hack/docs/metrics_gen_docs.go

@@ -56,6 +56,14 @@ func main() {
 		packages := getPackages(flag.Arg(i))
 		allMetrics = append(allMetrics, getMetricsFromPackages(packages...)...)
 	}
+	// Controller Runtime naming is different in that they don't specify a namespace or subsystem
+	// Getting the metrics requires special parsing logic
+	for i := range allMetrics {
+		if allMetrics[i].subsystem == "" && strings.HasPrefix(allMetrics[i].name, "controller_runtime_") {
+			allMetrics[i].subsystem = "controller_runtime"
+			allMetrics[i].name = strings.TrimPrefix(allMetrics[i].name, "controller_runtime_")
+		}
+	}
 	sort.Slice(allMetrics, bySubsystem(allMetrics))
 
 	outputFileName := flag.Arg(flag.NArg() - 1)
@@ -80,19 +88,15 @@ description: >
 	previousSubsystem := ""
 
 	for _, metric := range allMetrics {
-		// Controller Runtime naming is different in that they don't specify a namespace or subsystem
-		// Getting the metrics requires special parsing logic
-		if metric.subsystem == "" && strings.HasPrefix(metric.name, "controller_runtime_") {
-			metric.subsystem = "controller_runtime"
-			metric.name = strings.TrimPrefix(metric.name, "controller_runtime_")
-		}
 		if metric.subsystem != previousSubsystem {
-			subsystemTitle := strings.Join(lo.Map(strings.Split(metric.subsystem, "_"), func(s string, _ int) string {
-				return fmt.Sprintf("%s%s", strings.ToTitle(s[0:1]), s[1:])
-			}), " ")
-			fmt.Fprintf(f, "## %s Metrics\n", subsystemTitle)
+			if metric.subsystem != "" {
+				subsystemTitle := strings.Join(lo.Map(strings.Split(metric.subsystem, "_"), func(s string, _ int) string {
+					return fmt.Sprintf("%s%s", strings.ToTitle(s[0:1]), s[1:])
+				}), " ")
+				fmt.Fprintf(f, "## %s Metrics\n", subsystemTitle)
+				fmt.Fprintln(f)
+			}
 			previousSubsystem = metric.subsystem
-			fmt.Fprintln(f)
 		}
 		fmt.Fprintf(f, "### `%s`\n", metric.qualifiedName())
 		fmt.Fprintf(f, "%s\n", metric.help)
@@ -155,19 +159,23 @@ func getMetricsFromPackages(packages ...*ast.Package) []metricInfo {
 }
 
 func bySubsystem(metrics []metricInfo) func(i int, j int) bool {
-	subSystemSortOrder := map[string]int{}
-	subSystemSortOrder["nodepool"] = 1
-	subSystemSortOrder["nodes"] = 2
-	subSystemSortOrder["pods"] = 3
-	subSystemSortOrder["cloudprovider"] = 4
-	subSystemSortOrder["cloudprovider_batcher"] = 5
+	// Higher ordering comes first. If a value isn't designated here then the subsystem will be given a default of 0.
+	// Metrics without a subsystem come first since there is no designation for the bucket they fall under
+	subSystemSortOrder := map[string]int{
+		"":          100,
+		"nodepool":  10,
+		"nodeclaim": 9,
+		"nodes":     8,
+		"pods":      7,
+	}
+
 	return func(i, j int) bool {
 		lhs := metrics[i]
 		rhs := metrics[j]
 		if subSystemSortOrder[lhs.subsystem] != subSystemSortOrder[rhs.subsystem] {
-			return subSystemSortOrder[lhs.subsystem] < subSystemSortOrder[rhs.subsystem]
+			return subSystemSortOrder[lhs.subsystem] > subSystemSortOrder[rhs.subsystem]
 		}
-		return lhs.qualifiedName() < rhs.qualifiedName()
+		return lhs.qualifiedName() > rhs.qualifiedName()
 	}
 }
 

pkg/operator/operator.go

@@ -77,7 +77,7 @@ var BuildInfo = prometheus.NewGaugeVec(
 		Name:      "build_info",
 		Help:      "A metric with a constant '1' value labeled by version from which karpenter was built.",
 	},
-	[]string{"version", "goversion", "sha"})
+	[]string{"version", "goversion", "commit"})
 
 func init() {
 	lo.Must0(apis.AddToScheme(scheme.Scheme))

pkg/operator/suite_test.go

@@ -22,15 +22,16 @@ import (
 	"github.com/aws/aws-sdk-go/service/eks"
 	"github.com/samber/lo"
 
+	prometheusmodel "github.com/prometheus/client_model/go"
+	"sigs.k8s.io/karpenter/pkg/operator/scheme"
+	coretest "sigs.k8s.io/karpenter/pkg/test"
+
 	"github.com/aws/karpenter-provider-aws/pkg/apis"
 	"github.com/aws/karpenter-provider-aws/pkg/fake"
 	awscontext "github.com/aws/karpenter-provider-aws/pkg/operator"
 	"github.com/aws/karpenter-provider-aws/pkg/operator/options"
 	"github.com/aws/karpenter-provider-aws/pkg/test"
 
-	"sigs.k8s.io/karpenter/pkg/operator/scheme"
-	coretest "sigs.k8s.io/karpenter/pkg/test"
-
 	. "github.com/onsi/ginkgo/v2"
 	. "github.com/onsi/gomega"
 	. "knative.dev/pkg/logging/testing"
@@ -102,4 +103,13 @@ var _ = Describe("Operator", func() {
 		_, err := awscontext.ResolveClusterEndpoint(ctx, fakeEKSAPI)
 		Expect(err).To(HaveOccurred())
 	})
+	It("should fire a metric with the build_info", func() {
+		m, found := FindMetricWithLabelValues("karpenter_build_info", map[string]string{})
+		Expect(found).To(BeTrue())
+
+		for _, label := range []string{"version", "goversion", "commit"} {
+			_, ok := lo.Find(m.GetLabel(), func(l *prometheusmodel.LabelPair) bool { return lo.FromPtr(l.Name) == label })
+			Expect(ok).To(BeTrue())
+		}
+	})
 })