Add Virt Collectd plugin

This patch adds the virt collectd plugin to Browbeat.
The plugin monitors CPU, disk, network and memory resources on VMs
on compute nodes through information provided by the hypervisor, so
performance of the VMs is not majorly affected by the plugin.

Change-Id: I617cb3c29ae417c87d6b2c2652610ff0e8bcd4b1

ansible/install/group_vars/all.yml

@@ -409,6 +409,11 @@ osp_individual_resources_response_time: false
 # all metrics on all compute nodes
 ovn_compute_collectd_plugin: false
 
+#######################################################################
+# Virt plugin to monitor resources in VMs on computes
+#######################################################################
+virt_collectd_plugin: false
+
 ############################################
 # Prometheus/Collectd-Exporter Configuration
 ############################################

ansible/install/roles/collectd/tasks/main.yml

@@ -270,6 +270,9 @@
           {% if config_type == 'controller' and ceph_controller_collectd_plugin and inventory_hostname == groups['Controller'][0] %}
           -v /etc/ceph/:/etc/ceph/ -v /var/run/ceph/:/var/run/ceph \
           {% endif %}
+          {% if config_type == 'compute' and virt_collectd_plugin %}
+          -v /var/run/libvirt/libvirt-sock-ro:/var/run/libvirt/libvirt-sock-ro \
+          {% endif %}
           {% if config_type == 'cephstorage' and ceph_storage_collectd_plugin and inventory_hostname == groups['CephStorage'][0]  %}
           -v /var/run/ceph/:/var/run/ceph \
           {% endif %}

ansible/install/roles/collectd/templates/compute.collectd.conf.j2

@@ -48,6 +48,9 @@ LoadPlugin processes
 LoadPlugin swap
 LoadPlugin tail
 LoadPlugin uptime
+{% if virt_collectd_plugin %}
+LoadPlugin virt
+{% endif %}
 
 PreCacheChain "PreCache"
 <Chain "PreCache">
@@ -127,6 +130,15 @@ PreCacheChain "PreCache"
   ReportInodes true
 </Plugin>
 
+{% if virt_collectd_plugin %}
+<Plugin virt>
+  Connection "qemu:///system"
+  RefreshInterval 60
+  HostnameFormat name hostname uuid
+  ExtraStats "cpu_util disk job_stats_background"
+</Plugin>
+{% endif %}
+
 {%if disk_compute_collectd_plugin %}
 <Plugin disk>
   Disk "/^[hsv]d[a-z]+[0-9]?$/"
@@ -159,6 +171,7 @@ PreCacheChain "PreCache"
 </Plugin>
 
 {% endif %}
+
 # (akrzos) Including the version of OpenStack that the process was verified as running after
 # OpenStack Installation with a comment at the end of each Process/ProcessMatch statement.
 # A Minus before the version means the process was not found in that version. (Ex -10)

ansible/install/roles/grafana-dashboards/templates/openstack_general_system_performance.yaml.j2

@@ -177,6 +177,8 @@ dashboard:
 
 {% include 'partials/tail.yaml' %}
 
+{% include 'partials/virt_monitoring.yaml' %}
+
 {% if item.template_node_type in sqlalchemy_groups %}
   {% include 'partials/sqlalchemy.yaml' %}
 {% endif %}

ansible/install/roles/grafana-dashboards/templates/partials/virt_monitoring.yaml

@@ -0,0 +1,70 @@
+    - title: Virt Metrics
+      collapse: true
+      height: 200px
+      showTitle: true
+      panels:
+        - title: $Cloud - $Node - Virt Disk Metrics
+          type: graph
+          legend:
+            alignAsTable: true
+            avg: false
+            current: true
+            max: true
+            min: true
+            rightSide: true
+            show: true
+            total: false
+            values: true
+          nullPointMode: 'null'
+          targets:
+            - target: aliasByNode($Cloud.$Node.virt.disk*.*, 3, 4)
+            - target: aliasByMetric($Cloud.$Node.virt.total*)
+
+        - title: $Cloud - $Node - Virt Network Metrics
+          type: graph
+          legend:
+            alignAsTable: true
+            avg: false
+            current: true
+            max: true
+            min: true
+            rightSide: true
+            show: true
+            total: false
+            values: true
+          nullPointMode: 'null'
+          targets:
+            - target: aliasByNode($Cloud.$Node.virt.if*.*, 3, 4)
+
+        - title: $Cloud - $Node - Virt Memory Metrics
+          type: graph
+          legend:
+            alignAsTable: true
+            avg: false
+            current: true
+            max: true
+            min: true
+            rightSide: true
+            show: true
+            total: false
+            values: true
+          nullPointMode: 'null'
+          targets:
+            - target: alias($Cloud.$Node.virt.memory-total, 'memory-total')
+
+        - title: $Cloud - $Node - Virt vCPU Metrics
+          type: graph
+          legend:
+            alignAsTable: true
+            avg: false
+            current: true
+            max: true
+            min: true
+            rightSide: true
+            show: true
+            total: false
+            values: true
+          nullPointMode: 'null'
+          targets:
+            - target: alias($Cloud.$Node.virt.percent-virt_cpu_total, 'percent-virt_cpu_total')
+            - target: alias($Cloud.$Node.virt.virt_cpu_total, 'virt_cpu_total')

browbeat-containers/collectd-openstack/Dockerfile

@@ -5,7 +5,7 @@ RUN dnf clean all && \
     dnf install -y https://dl.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm && \
     dnf install -y centos-release-opstools && \
     dnf install -y collectd collectd-turbostat collectd-disk collectd-apache collectd-ceph \
-     collectd-mysql collectd-python collectd-ping python3-sqlalchemy-collectd && \
+     collectd-mysql collectd-python collectd-ping collectd-virt python3-sqlalchemy-collectd && \
     dnf install -y sysstat && \
     dnf install -y python3-pip python3-devel && \
     pip3 install --upgrade pip && \
@@ -22,7 +22,7 @@ RUN useradd stack
 RUN echo stack | passwd stack --stdin
 RUN echo "stack ALL=(root) NOPASSWD:ALL" | tee -a /etc/sudoers.d/stack
 RUN chmod 0440 /etc/sudoers.d/stack
-
+RUN rm /etc/collectd.d/virt.conf
 
 ADD files/collectd_ceph_storage.py /usr/local/bin/collectd_ceph_storage.py
 ADD files/collectd_gnocchi_status.py /usr/local/bin/collectd_gnocchi_status.py