Gather some error messages from the kube-controller-manager containers (#596)

Author: tremes

docs/gathered-data.md

@@ -383,6 +383,22 @@ installed in the cluster
   * 4.10+
 
 
+## KubeControllerManagerLogs
+
+collects logs from kube-controller-manager pods in the openshift-kube-controller-manager namespace with following substrings:
+  - "Internal error occurred: error resolving resource",
+  - "syncing garbage collector with updated resources from discovery",
+
+The Kubernetes API:
+         https://github.com/kubernetes/client-go/blob/master/kubernetes/typed/core/v1/pod_expansion.go#L48
+Response see:
+         https://docs.openshift.com/container-platform/4.10/rest_api/workloads_apis/pod-v1.html#apiv1namespacesnamespacepodsnamelog
+
+* Location in archive: config/pod/openshift-kube-controller-manager/logs/{pod-name}/errors.log
+* Since versions:
+  * 4.11+
+
+
 ## LogsOfNamespace
 
 creates a gathering closure which collects logs from pods in the provided namespace

docs/insights-archive-sample/config/pod/openshift-kube-controller-manager/logs/kube-controller-manager-ip-10-0-168-11.us-east-2.compute.internal/errors.log

@@ -0,0 +1,2 @@
+2022-03-16T13:32:39.044141390Z I0316 13:32:39.044104       1 garbagecollector.go:210] syncing garbage collector with updated resources from discovery (attempt 1): added: [], removed: [apps.openshift.io/v1, Resource=deploymentconfigs authorization.openshift.io/v1, Resource=rolebindingrestrictions build.openshift.io/v1, Resource=buildconfigs build.openshift.io/v1, Resource=builds image.openshift.io/v1, Resource=images image.openshift.io/v1, Resource=imagestreams project.openshift.io/v1, Resource=projects quota.openshift.io/v1, Resource=clusterresourcequotas route.openshift.io/v1, Resource=routes security.openshift.io/v1, Resource=rangeallocations security.openshift.io/v1, Resource=securitycontextconstraints template.openshift.io/v1, Resource=brokertemplateinstances template.openshift.io/v1, Resource=templateinstances template.openshift.io/v1, Resource=templates]
+2022-03-16T13:38:09.379847879Z I0316 13:38:09.379813       1 garbagecollector.go:210] syncing garbage collector with updated resources from discovery (attempt 1): added: [apps.openshift.io/v1, Resource=deploymentconfigs authorization.openshift.io/v1, Resource=rolebindingrestrictions build.openshift.io/v1, Resource=buildconfigs build.openshift.io/v1, Resource=builds image.openshift.io/v1, Resource=images image.openshift.io/v1, Resource=imagestreams project.openshift.io/v1, Resource=projects quota.openshift.io/v1, Resource=clusterresourcequotas route.openshift.io/v1, Resource=routes security.openshift.io/v1, Resource=rangeallocations security.openshift.io/v1, Resource=securitycontextconstraints template.openshift.io/v1, Resource=brokertemplateinstances template.openshift.io/v1, Resource=templateinstances template.openshift.io/v1, Resource=templates], removed: []

pkg/gatherers/clusterconfig/clusterconfig_gatherer.go

@@ -77,6 +77,7 @@ var gatheringFunctions = map[string]gathererFuncPtr{
 	"scheduler_logs":                    (*Gatherer).GatherSchedulerLogs,
 	"silenced_alerts":                   (*Gatherer).GatherSilencedAlerts,
 	"image":                             (*Gatherer).GatherClusterImage,
+	"kube_controller_manager_logs":      (*Gatherer).GatherKubeControllerManagerLogs,
 }
 
 func New(

pkg/gatherers/clusterconfig/kube_controller_manager_logs.go

@@ -0,0 +1,60 @@
+//nolint: dupl
+package clusterconfig
+
+import (
+	"context"
+
+	"k8s.io/client-go/kubernetes"
+
+	"github.com/openshift/insights-operator/pkg/gatherers/common"
+	"github.com/openshift/insights-operator/pkg/record"
+)
+
+// GatherKubeControllerManagerLogs collects logs from kube-controller-manager pods in the openshift-kube-controller-manager namespace with following substrings:
+//   - "Internal error occurred: error resolving resource",
+//   - "syncing garbage collector with updated resources from discovery",
+//
+// The Kubernetes API:
+//          https://github.com/kubernetes/client-go/blob/master/kubernetes/typed/core/v1/pod_expansion.go#L48
+// Response see:
+//          https://docs.openshift.com/container-platform/4.10/rest_api/workloads_apis/pod-v1.html#apiv1namespacesnamespacepodsnamelog
+//
+// * Location in archive: config/pod/openshift-kube-controller-manager/logs/{pod-name}/errors.log
+// * Since versions:
+//   * 4.11+
+func (g *Gatherer) GatherKubeControllerManagerLogs(ctx context.Context) ([]record.Record, []error) {
+	containersFilter := common.LogContainersFilter{
+		Namespace:                "openshift-kube-controller-manager",
+		LabelSelector:            "app=kube-controller-manager",
+		ContainerNameRegexFilter: "kube-controller-manager",
+	}
+	messagesFilter := common.LogMessagesFilter{
+		MessagesToSearch: []string{
+			"Internal error occurred: error resolving resource",
+			"syncing garbage collector with updated resources from discovery",
+		},
+		IsRegexSearch: true,
+		SinceSeconds:  logDefaultSinceSeconds,
+		LimitBytes:    0,
+	}
+
+	gatherKubeClient, err := kubernetes.NewForConfig(g.gatherProtoKubeConfig)
+	if err != nil {
+		return nil, []error{err}
+	}
+
+	coreClient := gatherKubeClient.CoreV1()
+
+	records, err := common.CollectLogsFromContainers(
+		ctx,
+		coreClient,
+		containersFilter,
+		messagesFilter,
+		nil,
+	)
+	if err != nil {
+		return nil, []error{err}
+	}
+
+	return records, nil
+}