Merged in phstsai/node-exporter-readme (pull request #3)

Phstsai/node exporter readme

Approved-by: Che Wei Lin <[email protected]>
Approved-by: Hung-Wei Chiu <[email protected]>

Former-commit-id: fec42ca552284137331d26f35a37c4596b5c9794 [formerly fec42ca552284137331d26f35a37c4596b5c9794 [formerly 3036a2e]]
Former-commit-id: 51d564d0dfbf48bdfd7dfe6b8f93eadd7ba52e24
Former-commit-id: 869f7ac

deploy/monitoring/README.md

@@ -0,0 +1,57 @@
+# Monitoring
+
+## Prometheus
+Prometheus is a systems and service monitoring system. It collects metrics from configured targets at given intervals, evaluates rule expressions, displays the results, and can trigger alerts if some condition is observed to be true.
+
+### Create a monitoring namespace
+
+``` bash
+kubectl create -f monitoring-namespace.yaml
+```
+
+### ConfigMap for Prometheus
+It collects metrics from ```kubernetes-apiservers```, ```kubernetes-cadvisor```, ```kubernetes-nodes```, ```kubernetes-pods```(every node-exporter), ```kubernetes-service-endpoints``` by default.
+
+``` bash
+kubectl create -f prometheus-config.yaml
+```
+
+### Create the deployment and service
+
+``` bash
+kubectl create -f prometheus-deployment.yaml
+kubectl create -f prometheus-svc.yaml
+```
+
+### Use the Prometheus
+Open the browser and connect to ```hostIP:30003```, you can see whether the Prometheus collect the metrics from targets correctly in ```Status->Targets```.
+
+You also can enter a PromQL expression to query the data and show them in graph.
+
+``` SQL
+sum by(pod_name)(container_memory_usage_bytes{namespace="kube-system"})
+```
+
+Or pass the expression through url and get the json.
+``` bash
+http://hostIP:30003/api/v1/query?query=sum%20by(pod_name)(container_memory_usage_bytes%7Bnamespace%3D%22kube-system%22%7D)
+```
+
+
+## Node Exporter
+
+Node exporter for machine metrics, written in Go with pluggable metric collectors.
+
+### Create the node-exporter daemonset
+
+``` bash
+kubectl create -f node-exporter.yaml
+```
+
+If you want to check whether the daemonset is created successfully, you can create a service to expose the node-exporter.
+
+``` bash
+kubectl create -f node-exporter-svc.yaml
+``` 
+
+Open the browser and connect to ```hostIP:30910/metrics```, you can see the metrics of the host node (service will route the request to the pod in round robin order).

deploy/monitoring/node-exporter-svc.yaml

@@ -0,0 +1,14 @@
+kind: Service
+apiVersion: v1
+metadata:
+  namespace: monitoring
+  name: node-exporter
+  labels:
+    app: node-exporter
+spec:
+  type: NodePort
+  ports:
+  - port: 9100
+    nodePort: 30910
+  selector:
+    name: node-exporter

deploy/monitoring/node-exporter.yaml

@@ -2,20 +2,20 @@ apiVersion: extensions/v1beta1
 kind: DaemonSet
 metadata:
   namespace: monitoring
-  name: prom-node-exporter
+  name: node-exporter
 spec:
   template:
     metadata:
-      name:  prom-node-exporter
+      name:  node-exporter
       labels:
-        name: prom-node-exporter
+        name: node-exporter
       annotations:
         prometheus.io.scrape: "true"
     spec:
       hostPID: true
       containers:
-      - name:  prom-node-exporter
-        image: prom/node-exporter
+      - name:  node-exporter
+        image: asia.gcr.io/linker-aurora/node-exporter:develop
         securityContext:
           privileged: true
         ports: