Merge pull request #1372 from stenwt/sturpin-2016-02-09-registry-health

added nodes not ready and nodes not schedulable triggers

roles/os_zabbix/vars/template_openshift_master.yml

@@ -201,6 +201,18 @@ g_template_openshift_master:
     applications:
     - Openshift Master Metrics
 
+  - key: openshift.master.nodesnotready.count
+    description: "This check shows how many nodes in a cluster are in NotReady state."
+    type: int
+    applications: 
+    - Openshift Master
+
+  - key: openshift.master.nodesnotschedulable.count
+    description: "This check shows how many nodes in a cluster are not schedulable."
+    type: int
+    applications:
+    - Openshift Master
+
   - key: openshift.master.apiserver.latency.summary.pods.quantile.list.5
     description: "Value from https://master/metrics.  This is the time, in miliseconds, that 50% of the pod operations have taken to completed."
     type: int
@@ -361,6 +373,20 @@ g_template_openshift_master:
     - 'Openshift Master API health check is failing on {HOST.NAME}'
     priority: high
 
+  - name: 'Hosts not ready according to {HOST.NAME}'
+    expression: '{Template Openshift Master:openshift.master.nodesnotready.count.last(#2)>0'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_node.asciidoc'
+    dependencies: 
+    - 'Openshift Master procecss not running on {HOST.NAME}'
+    priority: high
+
+  - name: 'Hosts not schedulable according to {HOST.NAME}'
+    expression: '{Template Openshift Master:openshift.master.nodesnotschedulable.count.last(#2)>0'
+    url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/openshift_node.asciidoc'
+    dependencies: 
+    - 'Openshift Master procecss not running on {HOST.NAME}'
+    priority: info
+
   zgraphs:
   - name: Openshift Master API Server Latency Pods LIST Quantiles
     width: 900