Gnocchi Performance Dashboard

Displays everything related to Gnocchi for Performance:
* Gnocchi Backlog against various metrics
* Gnocchi API Performance
* Gnocchi Process Metrics
* Ceilometer Process Metrics
* Specific Controller Process Metrics (Httpd, MariaDB, Rabbitmq, Mongodb, Redis)
* CephStorage Process Metrics
* Swift (Controller) Process Metrics
* Swift (ObjectStorage) Process Metrics
* Apache Metrics
* Rabbitmq Metrics
* Ceph Metrics
* Swift Metrics
* Gnocchi/Ceilometer/Ceph/MariaDB/Rabbitmq/Swift Log rates (E/W/I)

+ keeplastvalue on Instance dashboard to stop the "teardrop" affect on latest data

Change-Id: Id88bf96625396f81788c3ffeba90c89e0690b732

ansible/install/grafana-dashboards.yml

@@ -22,6 +22,7 @@
       - cloud_rabbitmq_monitoring
       - cloud_system_performance_comparsion
       - cloud_total_memory_usage
+      - gnocchi_performance
       - iostat
       - three_node_performance_food_groups
       - openstack_ironic_metrics

ansible/install/roles/grafana-dashboards/files/cloud_instance_count.json

@@ -49,7 +49,7 @@
             "targets": [
               {
                 "refId": "A",
-                "target": "alias(sumSeries($Cloud.*.processes-qemu-kvm.ps_count.processes), 'Sum qemu-kvm processes')",
+                "target": "alias(sumSeries(keepLastValue($Cloud.*.processes-qemu-kvm.ps_count.processes, 100)), 'Sum qemu-kvm processes')",
                 "textEditor": false
               }
             ],
@@ -124,7 +124,7 @@
             "targets": [
               {
                 "refId": "A",
-                "target": "aliasByNode($Cloud.*.processes-qemu-kvm.ps_count.processes, 1)",
+                "target": "aliasByNode(keepLastValue($Cloud.*.processes-qemu-kvm.ps_count.processes, 100), 1)",
                 "textEditor": false
               }
             ],

ansible/install/roles/grafana-dashboards/fix-ids.py

@@ -14,6 +14,15 @@ import argparse
 import json
 
 
+# Each Query in a Grafana Panel has a Reference ID, and this helps correct any drift or cut & paste
+# errors in the each Query's ID.
+refIds = [
+    "A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N",
+    "O", "P", "Q", "R", "S", "T", "U", "V", "W", "X", "Y", "Z", "AA", "AB",
+    "AC", "AD", "AE", "AF", "AG", "AH", "AI", "AJ", "AK", "AL", "AM", "AN",
+    "AO", "AP", "AQ", "AR", "AS", "AT", "AU", "AV", "AW", "AX", "AY", "AZ"
+    ]
+
 def main():
     """Script used to fix panel ids in static dashboards.  Typically adding a new panel or row into
     a static dashboard will involve re-ordering all subsequent panels.  This script automates that.
@@ -34,6 +43,15 @@ def main():
                 print "Found error in panel({}): {}".format(index, panel['title'])
                 panel['id'] = index
 
+            # Check over Query RefIds while after fixing Panel ID
+            refid_index = 0
+            for target in panel['targets']:
+                if target["refId"] != refIds[refid_index]:
+                    print "Panel: {}, Incorrect Target ID: {} Should be: {}".format(
+                        panel['title'], target["refId"], refIds[refid_index])
+                    target['refId'] = refIds[refid_index]
+                refid_index += 1
+
     with open(args.outputfile, 'w') as outputfile:
         json.dump(data, outputfile, sort_keys=True, indent=2, separators=(',', ': '))