CFME-collectd/dashbaord fixes.

+ Correctly collect on evm:dbsync:replicate, and appliance_console.rb + Fix Ansible 2.0 depreciation + Improve visualization of IOPs/Throughput per cfme process component of dashboard + Improve visualization of CFME Per Process CPU Utilization, 100% represents 100% cpu of a single core rather than all cpu resources. Change-Id: I35b5419c32c03e8f8cf1b1c7667ab9b1597997d8
2016-03-28 08:34:29 -04:00 · 2016-03-28 08:34:29 -04:00 · 121525e4e6
commit 121525e4e6
parent 8aeb443df9
7 changed files with 20 additions and 30 deletions
--- a/ansible/install/roles/collectd-generic/templates/cfme-all-in-one.collectd.conf.j2
+++ b/ansible/install/roles/collectd-generic/templates/cfme-all-in-one.collectd.conf.j2
@ -185,8 +185,8 @@ LoadPlugin uptime
  # Appliance processes
  ProcessMatch "evm_server.rb" ".*evm_server\.rb$"
  ProcessMatch "evm_watchdog.rb" "/bin/evm_watchdog.rb"
-  Process "evm:dbsync:replicate"
-  Process "appliance_console.rb"
+  ProcessMatch "evm:dbsync:replicate" ".*evm:dbsync:replicate$"
+  ProcessMatch "appliance_console.rb" ".*appliance_console.rb$"

  # Workers
  ProcessMatch "MiqEventHandler" ".*MiqEventHandler.*"
--- a/ansible/install/roles/collectd-generic/templates/cfme-vmdb.collectd.conf.j2
+++ b/ansible/install/roles/collectd-generic/templates/cfme-vmdb.collectd.conf.j2
@ -185,8 +185,8 @@ LoadPlugin uptime
  # Appliance processes
  ProcessMatch "evm_server.rb" ".*evm_server\.rb$"
  ProcessMatch "evm_watchdog.rb" "/bin/evm_watchdog.rb"
-  Process "evm:dbsync:replicate"
-  Process "appliance_console.rb"
+  ProcessMatch "evm:dbsync:replicate" ".*evm:dbsync:replicate$"
+  ProcessMatch "appliance_console.rb" ".*appliance_console.rb$"

  # Workers
  ProcessMatch "MiqEventHandler" ".*MiqEventHandler.*"
--- a/ansible/install/roles/collectd-generic/templates/cfme.collectd.conf.j2
+++ b/ansible/install/roles/collectd-generic/templates/cfme.collectd.conf.j2
@ -67,8 +67,8 @@ LoadPlugin uptime
  # Appliance processes
  ProcessMatch "evm_server.rb" ".*evm_server\.rb$"
  ProcessMatch "evm_watchdog.rb" "/bin/evm_watchdog.rb"
-  Process "evm:dbsync:replicate"
-  Process "appliance_console.rb"
+  ProcessMatch "evm:dbsync:replicate" ".*evm:dbsync:replicate$"
+  ProcessMatch "appliance_console.rb" ".*appliance_console.rb$"

  # Workers
  ProcessMatch "MiqEventHandler" ".*MiqEventHandler.*"
--- a/ansible/install/roles/dashboard-generic/tasks/main.yml
+++ b/ansible/install/roles/dashboard-generic/tasks/main.yml
@ -6,7 +6,7 @@
 - name: Remove existing dashboards
  command: "curl -X DELETE -H 'Content-Type: application/json' http://{{grafana_username}}:{{grafana_password}}@{{grafana_host}}:{{grafana_port}}/api/dashboards/db/{{item.process_list_name|lower}}-general-system-performance"
  when: overwrite_existing
-  with_items: dashboards
+  with_items: "{{dashboards}}"

 - name: Ensure {{role_path}}/files directory exists
  file: path={{role_path}}/files state=directory
@ -15,12 +15,12 @@
  template:
    src: "{{item.template_name}}_general_system_performance.json.j2"
    dest: "{{role_path}}/files/{{item.process_list_name}}_general_system_performance.json"
-  with_items: dashboards
+  with_items: "{{dashboards}}"

 - name: Upload dashboards to grafana
  command: "curl -X POST -H 'Content-Type: application/json' -d @{{role_path}}/files/{{item.process_list_name}}_general_system_performance.json http://{{grafana_username}}:{{grafana_password}}@{{grafana_host}}:{{grafana_port}}/api/dashboards/db"
-  with_items: dashboards
+  with_items: "{{dashboards}}"

 - name: Remove leftover json file(s)
  file: path={{role_path}}/files/{{item.process_list_name}}_general_system_performance.json state=absent
-  with_items: dashboards
+  with_items: "{{dashboards}}"
--- a/ansible/install/roles/dashboard-generic/templates/cfme_general_system_performance.json.j2
+++ b/ansible/install/roles/dashboard-generic/templates/cfme_general_system_performance.json.j2
@ -1890,7 +1890,7 @@
            "lines": true,
            "linewidth": 2,
            "links": [],
-            "nullPointMode": "connected",
+            "nullPointMode": "{{metrics.nullPointMode}}",
            "percentage": false,
            "pointradius": 5,
            "points": false,
@ -1901,20 +1901,12 @@
            "steppedLine": false,
            "targets": [
        {% if vars.update({'temp_count': 0}) %} {% endif %}
-        {% if metrics.name|match('Process CPU') %}
-              {
-                "hide": true,
-                "refId": "{{vars.temp_count|to_grafana_refid}}",
-                "target": "alias(sumSeries(offset(scale($Cloud.$Node.cpu-*.cpu-idle, 0), 1)), 'cores')"
-              },
-              {% if vars.update({'temp_count': (vars.temp_count + 1)}) %} {% endif %}
-        {% endif %}
        {% for process in panel.processes %}
          {% for metric in metrics.metrics %}
              {
                "refId": "{{vars.temp_count|to_grafana_refid}}",
            {% if metrics.name|match('Process CPU') %}
-                "target": "alias(divideSeries(scale($Cloud.$Node.processes-{{process|replace('.', '_')|replace(':', '_')}}{{metric.query}}, 0.0001), #A), '{{process}} {{metric.name}}')"
+                "target": "alias(scale($Cloud.$Node.processes-{{process|replace('.', '_')|replace(':', '_')}}{{metric.query}}, 0.0001), '{{process}} {{metric.name}}')"
            {% else %}
                "target": "alias($Cloud.$Node.processes-{{process|replace('.', '_')|replace(':', '_')}}{{metric.query}}, '{{process}} {{metric.name}}')"
            {% endif %}
--- a/ansible/install/roles/dashboard-generic/templates/cfmeallinone_general_system_performance.json.j2
+++ b/ansible/install/roles/dashboard-generic/templates/cfmeallinone_general_system_performance.json.j2
@ -1890,7 +1890,7 @@
            "lines": true,
            "linewidth": 2,
            "links": [],
-            "nullPointMode": "connected",
+            "nullPointMode": "{{metrics.nullPointMode}}",
            "percentage": false,
            "pointradius": 5,
            "points": false,
@ -1901,20 +1901,12 @@
            "steppedLine": false,
            "targets": [
        {% if vars.update({'temp_count': 0}) %} {% endif %}
-        {% if metrics.name|match('Process CPU') %}
-              {
-                "hide": true,
-                "refId": "{{vars.temp_count|to_grafana_refid}}",
-                "target": "alias(sumSeries(offset(scale($Cloud.$Node.cpu-*.cpu-idle, 0), 1)), 'cores')"
-              },
-              {% if vars.update({'temp_count': (vars.temp_count + 1)}) %} {% endif %}
-        {% endif %}
        {% for process in panel.processes %}
          {% for metric in metrics.metrics %}
              {
                "refId": "{{vars.temp_count|to_grafana_refid}}",
            {% if metrics.name|match('Process CPU') %}
-                "target": "alias(divideSeries(scale($Cloud.$Node.processes-{{process|replace('.', '_')|replace(':', '_')}}{{metric.query}}, 0.0001), #A), '{{process}} {{metric.name}}')"
+                "target": "alias(scale($Cloud.$Node.processes-{{process|replace('.', '_')|replace(':', '_')}}{{metric.query}}, 0.0001), '{{process}} {{metric.name}}')"
            {% else %}
                "target": "alias($Cloud.$Node.processes-{{process|replace('.', '_')|replace(':', '_')}}{{metric.query}}, '{{process}} {{metric.name}}')"
            {% endif %}
--- a/ansible/install/roles/dashboard-generic/vars/main.yml
+++ b/ansible/install/roles/dashboard-generic/vars/main.yml
@ -11,6 +11,7 @@ per_process_metrics:
        query: ".ps_count.processes"
      - name: "Threads"
        query: ".ps_count.threads"
+    nullPointMode: "connected"
  - name: "Process CPU"
    y1units: "percent"
    metrics:
@ -18,6 +19,7 @@ per_process_metrics:
        query: ".ps_cputime.syst"
      - name: "User"
        query: ".ps_cputime.user"
+    nullPointMode: "connected"
  - name: "Process Memory"
    y1units: "bits"
    metrics:
@ -25,6 +27,7 @@ per_process_metrics:
        query: ".ps_rss"
      - name: "Virtual"
        query: ".ps_vm"
+    nullPointMode: "connected"
  - name: "Process Page Faults"
    y1units: "short"
    metrics:
@ -32,6 +35,7 @@ per_process_metrics:
        query: ".ps_pagefaults.majflt"
      - name: "Minflt"
        query: ".ps_pagefaults.minflt"
+    nullPointMode: "connected"
  - name: "Process IOPs(Estimated via SYSCALLS)"
    y1units: "iops"
    metrics:
@ -39,6 +43,7 @@ per_process_metrics:
        query: ".ps_disk_ops.read"
      - name: "Write"
        query: ".ps_disk_ops.write"
+    nullPointMode: "null"
  - name: "Process IO Throughput(Estimated via SYSCALLS)"
    y1units: "bytes"
    metrics:
@ -46,6 +51,7 @@ per_process_metrics:
        query: ".ps_disk_octets.read"
      - name: "Write"
        query: ".ps_disk_octets.write"
+    nullPointMode: "null"

 per_process_panels:
  Baremetal: