Fix K8s monitoring the incorrect metrics calculates. (#6739)

CHANGES.md

@@ -11,7 +11,7 @@ Release Notes.
 
 
 #### OAP-Backend
-
+* Fix K8s monitoring the incorrect metrics calculate. 
 
 #### UI
 

oap-server/server-bootstrap/src/main/resources/otel-oc-rules/k8s-node.yaml

@@ -36,7 +36,7 @@ metricsRules:
   - name: cpu_cores
     exp: (kube_node_status_capacity * 1000).tagEqual('resource' , 'cpu').sum(['cluster' , 'node'])
   - name: cpu_usage
-    exp: (container_cpu_usage_seconds_total * 1000).tagEqual('id' , '/').sum(['cluster' , 'node']).rate('PT1M')
+    exp: (container_cpu_usage_seconds_total * 1000).tagEqual('id' , '/').sum(['cluster' , 'node']).irate()
   - name: cpu_cores_allocatable
     exp: (kube_node_status_allocatable * 1000).tagEqual('resource' , 'cpu').sum(['cluster' , 'node'])
   - name: cpu_cores_requests
@@ -69,6 +69,6 @@ metricsRules:
     exp: kube_pod_info.sum(['cluster' , 'node'])
 
   - name: network_receive
-    exp: container_network_receive_bytes_total.sum(['cluster' , 'node']).irate()
+    exp: container_network_receive_bytes_total.tagEqual('id' , '/').sum(['cluster' , 'node']).irate()
   - name: network_transmit
-    exp: container_network_transmit_bytes_total.sum(['cluster' , 'node']).irate()
+    exp: container_network_transmit_bytes_total.tagEqual('id' , '/').sum(['cluster' , 'node']).irate()

oap-server/server-bootstrap/src/main/resources/otel-oc-rules/k8s-service.yaml

@@ -54,13 +54,13 @@ metricsRules:
     exp: kube_pod_container_status_restarts_total.retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod'])
 
   - name: pod_cpu_usage
-    exp: (container_cpu_usage_seconds_total * 1000).tagNotEqual('pod' , '').retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod']).rate('PT1M')
+    exp: (container_cpu_usage_seconds_total * 1000).tagNotEqual('container' , '').tagNotEqual('pod' , '').retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod']).irate()
   - name: pod_memory_usage
-    exp: container_memory_working_set_bytes.retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod'])
+    exp: container_memory_working_set_bytes.tagNotEqual('container' , '').tagNotEqual('pod' , '').retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod'])
 
   - name: pod_network_receive
-    exp: container_network_receive_bytes_total.tagNotEqual('pod' , '').retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod']).irate()
+    exp: container_network_receive_bytes_total.tagNotEqual('container' , '').tagNotEqual('pod' , '').retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod']).irate()
   - name: pod_network_transmit
-    exp: container_network_transmit_bytes_total.tagNotEqual('pod' , '').retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod']).irate()
+    exp: container_network_transmit_bytes_total.tagNotEqual('container' , '').tagNotEqual('pod' , '').retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod']).irate()
   - name: pod_fs_usage
-    exp: container_fs_usage_bytes.tagNotEqual('pod' , '').retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod'])
+    exp: container_fs_usage_bytes.tagNotEqual('container' , '').tagNotEqual('pod' , '').retagByK8sMeta('service' , K8sRetagType.Pod2Service , 'pod' , 'namespace').tagNotEqual('service' , '').sum(['cluster' , 'service' , 'pod'])