Adding a replication controller for all monitoring pods.

Removed auth for Grafana to facilitate usage via service proxy on the api-server. Added a grafana service Removed elasticsearch dependency for monitoring - faster startup times.
2015-01-15 00:01:12 +00:00 · 2015-01-15 00:01:12 +00:00 · 50c32fbba4
parent efef4b7d5e
commit 50c32fbba4
9 changed files with 102 additions and 135 deletions
--- a/cluster/addons/cluster-monitoring/grafana-service.yaml
+++ b/cluster/addons/cluster-monitoring/grafana-service.yaml
@ -0,0 +1,7 @@
+apiVersion: "v1beta1"
+kind: "Service"
+id: "monitoring-grafana"
+port: 8087
+containerPort: 80
+selector: 
+  name: "influxGrafana"
--- a/cluster/addons/cluster-monitoring/heapster-controller.yaml
+++ b/cluster/addons/cluster-monitoring/heapster-controller.yaml
@ -0,0 +1,20 @@
+apiVersion: "v1beta1"
+id: "monitoring-heapsterController"
+kind: "ReplicationController"
+desiredState: 
+  replicas: 1
+  replicaSelector:    
+    name: "heapster"
+  podTemplate:
+    desiredState:
+      manifest:
+        version: "v1beta1"
+        id: "monitoring-heapsterController"
+        containers:
+          - name: "heapster"
+            image: "kubernetes/heapster:v0.5"
+    labels: 
+      name: "heapster"
+      uses: "monitoring-influxdb"
+labels: 
+  name: "heapster"
--- a/cluster/addons/cluster-monitoring/heapster-pod.json
+++ b/cluster/addons/cluster-monitoring/heapster-pod.json
@ -1,20 +0,0 @@
-{
-  "id": "heapster",
-  "kind": "Pod",
-  "apiVersion": "v1beta1",
-  "desiredState": {
-    "manifest": {
-      "version": "v1beta1",
-      "id": "heapster",
-      "containers": [
-        {
-          "name": "heapster",
-          "image": "kubernetes/heapster"
-        }
-      ]
-    }
-  },
-  "labels": {
-    "name": "heapster"
-  }
-}
--- a/cluster/addons/cluster-monitoring/influx-grafana-pod.json
+++ b/cluster/addons/cluster-monitoring/influx-grafana-pod.json
@ -1,71 +0,0 @@
-{
-  "id": "influx-grafana",
-  "kind": "Pod",
-  "apiVersion": "v1beta1",
-  "desiredState": {
-    "manifest": {
-      "version": "v1beta1",
-      "id": "influx-grafana",
-      "containers": [
-        {
-          "name": "influxdb",
-          "image": "kubernetes/heapster_influxdb",
-          "ports": [
-            {
-              "containerPort": 8083,
-              "hostPort": 8083
-            },
-            {
-              "containerPort": 8086,
-              "hostPort": 8086
-            },
-            {
-              "containerPort": 8090,
-              "hostPort": 8090
-            },
-            {
-              "containerPort": 8099,
-              "hostPort": 8099
-            }
-          ]
-        },
-        {
-          "name": "grafana",
-          "image": "kubernetes/heapster_grafana",
-          "ports": [
-            {
-              "containerPort": 80,
-              "hostPort": 80
-            }
-          ],
-          "env": [
-            {
-              "name": "HTTP_USER",
-              "value": "{KUBE_USER}"
-            },
-            {
-              "name": "HTTP_PASS",
-              "value": "{KUBE_PASSWORD}"
-            }
-          ]
-        },
-        {
-          "name": "elasticsearch",
-          "image": "dockerfile/elasticsearch",
-          "ports": [
-            {
-              "containerPort": 9200,
-              "hostPort": 9200
-            },
-            {
-              "containerPort": 9300
-            }
-          ]
-        }
-      ]
-    }
-  },
-  "labels": {
-    "name": "influxdb"
-  }
-}
--- a/cluster/addons/cluster-monitoring/influx-grafana-service.json
+++ b/cluster/addons/cluster-monitoring/influx-grafana-service.json
@ -1,12 +0,0 @@
-{
-  "id": "influx-master",
-  "kind": "Service",
-  "apiVersion": "v1beta1",
-  "port": 8085,
-  "containerPort": 8086,
-  "provider": "kubernetes-default",
-  "component": "influxdb",
-  "selector": {
-    "name": "influxdb"
-  }
-}
--- a/cluster/addons/cluster-monitoring/influxdb-grafana-controller.yaml
+++ b/cluster/addons/cluster-monitoring/influxdb-grafana-controller.yaml
@ -0,0 +1,34 @@
+apiVersion: "v1beta1"
+kind: "ReplicationController"
+id: "monitoring-influxGrafanaController"
+desiredState: 
+  replicas: 1
+  replicaSelector:
+    name: "influxGrafana"
+  podTemplate:
+    labels:
+      name: "influxGrafana"
+    desiredState:
+      manifest:
+        version: "v1beta1"
+        id: "monitoring-influxGrafanaController"
+        containers: 
+          - name: "influxdb"
+            image: "kubernetes/heapster_influxdb:v0.2"
+            ports: 
+              - containerPort: 8083
+                hostPort: 8083
+              - containerPort: 8086
+                hostPort: 8086
+          - name: "grafana"
+            image: "kubernetes/heapster_grafana:v0.2"
+            ports: 
+              - containerPort: 80
+                hostPort: 80
+            env: 
+              - name: "HTTP_USER"
+                value: "admin"
+              - name: "HTTP_PASS"
+                value: "**None**"
+labels:
+  name: "influxGrafana"
--- a/cluster/addons/cluster-monitoring/influxdb-service.yaml
+++ b/cluster/addons/cluster-monitoring/influxdb-service.yaml
@ -0,0 +1,7 @@
+apiVersion: "v1beta1"
+kind: "Service"
+id: "monitoring-influxdb"
+port: 8085
+containerPort: 8086
+selector: 
+  name: "influxGrafana"
--- a/cluster/gce/util.sh
+++ b/cluster/gce/util.sh
@ -737,35 +737,37 @@ function setup-monitoring {
    echo "Setting up cluster monitoring using Heapster."

    detect-project
-    if ! gcloud compute firewall-rules --project "{$PROJECT}" describe monitoring-heapster &>/dev/null; then
+    if ! gcloud compute firewall-rules --project "${PROJECT}" describe monitoring-heapster &> /dev/null; then
      if ! gcloud compute firewall-rules create monitoring-heapster \
          --project "${PROJECT}" \
          --target-tags="${MINION_TAG}" \
          --network="${NETWORK}" \
-          --allow tcp:80 tcp:8083 tcp:8086 tcp:9200; then
-        echo "Failed to set up firewall for monitoring" && false
+          --allow tcp:80 tcp:8083 tcp:8086; then
+        echo -e "${color_red}Failed to set up firewall for monitoring ${color_norm}" && false
      fi
    fi

-    # Re-use master auth for Grafana
-    get-password
-    ensure-temp-dir
-
-    sed -e "s/{KUBE_USER}/$KUBE_USER/g" \
-        -e "s/{KUBE_PASSWORD}/$KUBE_PASSWORD/g" \
-        "${KUBE_ROOT}/cluster/addons/cluster-monitoring/influx-grafana-pod.json" \
-      > "${KUBE_TEMP}/influx-grafana-pod.json"
    local kubectl="${KUBE_ROOT}/cluster/kubectl.sh"
-    if "${kubectl}" create -f "${KUBE_TEMP}/influx-grafana-pod.json" &> /dev/null \
-        && "${kubectl}" create -f "${KUBE_ROOT}/cluster/addons/cluster-monitoring/influx-grafana-service.json" &> /dev/null \
-        && "${kubectl}" create -f "${KUBE_ROOT}/cluster/addons/cluster-monitoring/heapster-pod.json" &> /dev/null; then
-      local dashboard_url="http://$(${kubectl} get -o json pod influx-grafana | grep hostIP | awk '{print $2}' | sed 's/[,|\"]//g')"
-      echo
-      echo "Grafana dashboard will be available at $dashboard_url. Wait for the monitoring dashboard to be online."
-      echo "Use the master user name and password for the dashboard."
-      echo
+    local grafana_host=""
+    if "${kubectl}" create -f "${KUBE_ROOT}/cluster/addons/cluster-monitoring/" &> /dev/null; then
+      # wait for pods to be scheduled on a node.
+      echo "waiting for monitoring pods to be scheduled."
+      for i in `seq 1 10`; do
+	grafana_host=$("${kubectl}" get pods -l name=influxGrafana -o template -t {{range.items}}{{.currentState.hostIP}}:{{end}} | sed s/://g) 
+	if [[ $grafana_host != *"<"* ]]; then
+	  break
+	fi
+	sleep 10
+      done
+      if [[ $grafana_host != *"<"* ]]; then
+	echo
+	echo -e "${color_green}Grafana dashboard will be available at ${color_yellow}http://$grafana_host${color_green}. Wait for the monitoring dashboard to be online.${color_norm}"
+	echo
+      else
+	echo -e "${color_red}monitoring pods failed to be scheduled.${color_norm}"
+      fi
    else
-      echo "Failed to Setup Monitoring"
+      echo -e "${color_red}Failed to Setup Monitoring ${color_norm}"
      teardown-monitoring
    fi
  fi
@ -776,9 +778,10 @@ function teardown-monitoring {
    detect-project

    local kubectl="${KUBE_ROOT}/cluster/kubectl.sh"
-    "${kubectl}" delete pods heapster &> /dev/null || true
-    "${kubectl}" delete pods influx-grafana &> /dev/null || true
-    "${kubectl}" delete services influx-master &> /dev/null || true
+    local kubecfg="${KUBE_ROOT}/cluster/kubecfg.sh"
+    "${kubecfg}" resize monitoring-influxGrafanaController 0 &> /dev/null || true
+    "${kubecfg}" resize monitoring-heapsterController 0 &> /dev/null || true
+    "${kubectl}" delete -f "${KUBE_ROOT}/cluster/addons/cluster-monitoring/" &> /dev/null || true
    if gcloud compute firewall-rules describe --project "${PROJECT}" monitoring-heapster &> /dev/null; then
      gcloud compute firewall-rules delete \
          --project "${PROJECT}" \
--- a/hack/e2e-suite/monitoring.sh
+++ b/hack/e2e-suite/monitoring.sh
@ -28,6 +28,7 @@ source "${KUBE_ROOT}/cluster/$KUBERNETES_PROVIDER/util.sh"

 MONITORING="${KUBE_ROOT}/cluster/addons/cluster-monitoring"
 KUBECTL="${KUBE_ROOT}/cluster/kubectl.sh"
+KUBECFG="${KUBE_ROOT}/cluster/kubecfg.sh"
 BIGRAND=$(printf "%x\n" $(( $RANDOM << 16 | $RANDOM ))) # random 2^32 in hex
 MONITORING_FIREWALL_RULE="monitoring-test-${BIGRAND}"

@ -44,15 +45,13 @@ function setup {
    fi
  fi

-  "${KUBECTL}" create -f "${MONITORING}/influx-grafana-pod.json"
-  "${KUBECTL}" create -f "${MONITORING}/influx-grafana-service.json"
-  "${KUBECTL}" create -f "${MONITORING}/heapster-pod.json"
+  "${KUBECTL}" create -f "${MONITORING}/"
 }

 function cleanup {
-  "${KUBECTL}" delete -f "${MONITORING}/influx-grafana-pod.json" || true
-  "${KUBECTL}" delete -f "${MONITORING}/influx-grafana-service.json" || true
-  "${KUBECTL}" delete -f "${MONITORING}/heapster-pod.json" || true
+  "${KUBECFG}" resize monitoring-influxGrafanaController 0 &> /dev/null || true
+  "${KUBECFG}" resize monitoring-heapsterController 0 &> /dev/null || true
+  "${KUBECTL}" delete -f "${MONITORING}/" &> /dev/null || true

  # This only has work to do on gce and gke
  if [[ "${KUBERNETES_PROVIDER}" == "gce" ]] || [[ "${KUBERNETES_PROVIDER}" == "gke" ]]; then
@ -69,7 +68,7 @@ function cleanup {
 function influx-data-exists {
  local max_retries=10
  local retry_delay=30 #seconds
-  local influx_ip=$("${KUBECTL}" get -o json pods influx-grafana | grep hostIP | awk '{print $2}' | sed 's/["|,]//g')
+  local influx_ip=$("${KUBECTL}" get pods -l name=influxGrafana -o template -t {{range.items}}{{.currentState.hostIP}}:{{end}} | sed s/://g)
  local influx_url="http://$influx_ip:8086/db/k8s/series?u=root&p=root"
  local ok="false"
  for i in `seq 1 10`; do
@ -91,8 +90,8 @@ function wait-for-pods {
  local running=false
  for i in `seq 1 20`; do
    sleep 20
-    if "${KUBECTL}" get pods influx-grafana | grep Running &> /dev/null \
-      && "${KUBECTL}" get pods heapster | grep Running &> /dev/null; then
+    if "${KUBECTL}" get pods -l name=influxGrafana -o template -t {{range.items}}{{.currentState.status}}:{{end}} | grep Running &> /dev/null \
+      && "${KUBECTL}" get pods -l name=heapster -o template -t {{range.items}}{{.currentState.status}}:{{end}} | grep Running &> /dev/null; then
      running=true
      break
    fi