Merge pull request #3186 from junotx/monitoring

support for querying pods metrics from the cluster level

pkg/kapis/monitoring/v1alpha3/register.go

@@ -219,6 +219,24 @@ func AddToContainer(c *restful.Container, k8sClient kubernetes.Interface, monito
 		Returns(http.StatusOK, respOK, model.Metrics{})).
 		Produces(restful.MIME_JSON)
 
+	ws.Route(ws.GET("/pods").
+		To(h.handlePodMetricsQuery).
+		Doc("Get pod-level metric data of the whole cluster's pods.").
+		Param(ws.QueryParameter("metrics_filter", "The metric name filter consists of a regexp pattern. It specifies which metric data to return. For example, the following filter matches both pod CPU usage and memory usage: `pod_cpu_usage|pod_memory_usage`. View available metrics at [kubesphere.io](https://docs.kubesphere.io/advanced-v2.0/zh-CN/api-reference/monitoring-metrics/).").DataType("string").Required(false)).
+		Param(ws.QueryParameter("resources_filter", "The pod filter consists of a regexp pattern. It specifies which pod data to return. For example, the following filter matches any pod whose name begins with redis: `redis.*`.").DataType("string").Required(false)).
+		Param(ws.QueryParameter("start", "Start time of query. Use **start** and **end** to retrieve metric data over a time span. It is a string with Unix time format, eg. 1559347200. ").DataType("string").Required(false)).
+		Param(ws.QueryParameter("end", "End time of query. Use **start** and **end** to retrieve metric data over a time span. It is a string with Unix time format, eg. 1561939200. ").DataType("string").Required(false)).
+		Param(ws.QueryParameter("step", "Time interval. Retrieve metric data at a fixed interval within the time range of start and end. It requires both **start** and **end** are provided. The format is [0-9]+[smhdwy]. Defaults to 10m (i.e. 10 min).").DataType("string").DefaultValue("10m").Required(false)).
+		Param(ws.QueryParameter("time", "A timestamp in Unix time format. Retrieve metric data at a single point in time. Defaults to now. Time and the combination of start, end, step are mutually exclusive.").DataType("string").Required(false)).
+		Param(ws.QueryParameter("sort_metric", "Sort pods by the specified metric. Not applicable if **start** and **end** are provided.").DataType("string").Required(false)).
+		Param(ws.QueryParameter("sort_type", "Sort order. One of asc, desc.").DefaultValue("desc.").DataType("string").Required(false)).
+		Param(ws.QueryParameter("page", "The page number. This field paginates result data of each metric, then returns a specific page. For example, setting **page** to 2 returns the second page. It only applies to sorted metric data.").DataType("integer").Required(false)).
+		Param(ws.QueryParameter("limit", "Page size, the maximum number of results in a single page. Defaults to 5.").DataType("integer").Required(false).DefaultValue("5")).
+		Metadata(restfulspec.KeyOpenAPITags, []string{constants.PodMetricsTag}).
+		Writes(model.Metrics{}).
+		Returns(http.StatusOK, respOK, model.Metrics{})).
+		Produces(restful.MIME_JSON)
+
 	ws.Route(ws.GET("/namespaces/{namespace}/pods").
 		To(h.handlePodMetricsQuery).
 		Doc("Get pod-level metric data of the specific namespace's pods.").

pkg/simple/client/monitoring/prometheus/promql.go

@@ -349,6 +349,12 @@ func makePodMetricExpr(tmpl string, o monitoring.QueryOptions) string {
 		}
 	}
 
+	// For monitoring pods in the whole cluster
+	// Get /pods
+	if o.NamespaceName == "" && o.NodeName == "" {
+		podSelector = fmt.Sprintf(`pod=~"%s"`, o.ResourceFilter)
+	}
+
 	// For monitoring pods in the specific namespace
 	// GET /namespaces/{namespace}/workloads/{kind}/{workload}/pods or
 	// GET /namespaces/{namespace}/pods/{pod} or

pkg/simple/client/monitoring/prometheus/promql_test.go

@@ -135,6 +135,13 @@ func TestMakeExpr(t *testing.T) {
 				PodName:  "elasticsearch-12345",
 			},
 		},
+		{
+			name: "pod_net_bytes_transmitted",
+			opts: monitoring.QueryOptions{
+				Level:          monitoring.LevelPod,
+				ResourceFilter: "elasticsearch-0",
+			},
+		},
 		{
 			name: "container_cpu_usage",
 			opts: monitoring.QueryOptions{

pkg/simple/client/monitoring/prometheus/testdata/promqls.go

@@ -31,6 +31,7 @@ var PromQLs = map[string]string{
 	"pod_cpu_usage":                         `round(sum by (namespace, pod) (irate(container_cpu_usage_seconds_total{job="kubelet", pod!="", image!=""}[5m])) * on (namespace, pod) group_left(owner_kind, owner_name) kube_pod_owner{owner_kind="ReplicaSet", owner_name=~"^elasticsearch-[^-]{1,10}$"} * on (namespace, pod) group_left(node) kube_pod_info{pod=~"elasticsearch-0", namespace="default"}, 0.001)`,
 	"pod_memory_usage":                      `sum by (namespace, pod) (container_memory_usage_bytes{job="kubelet", pod!="", image!=""}) * on (namespace, pod) group_left(owner_kind, owner_name) kube_pod_owner{} * on (namespace, pod) group_left(node) kube_pod_info{pod="elasticsearch-12345", namespace="default"}`,
 	"pod_memory_usage_wo_cache":             `sum by (namespace, pod) (container_memory_working_set_bytes{job="kubelet", pod!="", image!=""}) * on (namespace, pod) group_left(owner_kind, owner_name) kube_pod_owner{} * on (namespace, pod) group_left(node) kube_pod_info{pod="elasticsearch-12345", node="i-2dazc1d6"}`,
+	"pod_net_bytes_transmitted":             `sum by (namespace, pod) (irate(container_network_transmit_bytes_total{pod!="", interface!~"^(cali.+|tunl.+|dummy.+|kube.+|flannel.+|cni.+|docker.+|veth.+|lo.*)", job="kubelet"}[5m])) * on (namespace, pod) group_left(owner_kind, owner_name) kube_pod_owner{} * on (namespace, pod) group_left(node) kube_pod_info{pod=~"elasticsearch-0"}`,
 	"container_cpu_usage":                   `round(sum by (namespace, pod, container) (irate(container_cpu_usage_seconds_total{job="kubelet", container!="POD", container!="", image!="", pod="elasticsearch-12345", namespace="default", container="syscall"}[5m])), 0.001)`,
 	"container_memory_usage":                `sum by (namespace, pod, container) (container_memory_usage_bytes{job="kubelet", container!="POD", container!="", image!="", pod="elasticsearch-12345", namespace="default", container=~"syscall"})`,
 	"pvc_inodes_available":                  `max by (namespace, persistentvolumeclaim) (kubelet_volume_stats_inodes_free) * on (namespace, persistentvolumeclaim) group_left (storageclass) kube_persistentvolumeclaim_info{namespace="default", persistentvolumeclaim="db-123"}`,