Chapter 5. Configuring Red Hat OpenStack Platform director Operator for Service Telemetry Framework

Procedure

1. View a list of available certificates in STF:

   $ oc get secrets

2. Retrieve and note the content of the default-interconnect-selfsigned Secret:

   $ oc get secret/default-interconnect-selfsigned -o jsonpath='{.data.ca\.crt}' | base64 -d

Procedure

1. Log in to your Red Hat OpenShift Container Platform environment where STF is hosted.

2. Change to the service-telemetry project:

   $ oc project service-telemetry

3. Retrieve the AMQ Interconnect route address:

   $ oc get routes -ogo-template='{{ range .items }}{{printf "%s\n" .spec.host }}{{ end }}' | grep "\-5671"
   default-interconnect-5671-service-telemetry.apps.infra.watch

Procedure

1. Log in to the Red Hat OpenShift Container Platform environment where RHOSP director Operator is deployed and change to the project that hosts your RHOSP deployment:

   $ oc project openstack

2. Open the heat-env-config-deploy ConfigMap CR for editing:

   $ oc edit heat-env-config-deploy

3. Add the enable-stf.yaml configuration to the heat-env-config-deploy ConfigMap, save your edits and close the file:

   enable-stf.yaml

   apiVersion: v1
   data:
     [...]
     enable-stf.yaml: |
       parameter_defaults:
           # only send to STF, not other publishers
           PipelinePublishers: []
   
           # manage the polling and pipeline configuration files for Ceilometer agents
           ManagePolling: true
           ManagePipeline: true
           ManageEventPipeline: false
   
           # enable Ceilometer metrics and events
           CeilometerQdrPublishMetrics: true
   
           # enable collection of API status
           CollectdEnableSensubility: true
           CollectdSensubilityTransport: amqp1
   
           # enable collection of containerized service metrics
           CollectdEnableLibpodstats: true
   
           # set collectd overrides for higher telemetry resolution and extra plugins
           # to load
           CollectdConnectionType: amqp1
           CollectdAmqpInterval: 30
           CollectdDefaultPollingInterval: 30
           # to collect information about the virtual memory subsystem of the kernel
           # CollectdExtraPlugins:
           # - vmem
   
           # set standard prefixes for where metrics are published to QDR
           MetricsQdrAddresses:
           - prefix: 'collectd'
             distribution: multicast
           - prefix: 'anycast/ceilometer'
             distribution: multicast
   
           ExtraConfig:
              ceilometer::agent::polling::polling_interval: 30
              ceilometer::agent::polling::polling_meters:
              - cpu
              - memory.usage
   
              # to avoid filling the memory buffers if disconnected from the message bus
              # note: this may need an adjustment if there are many metrics to be sent.
              collectd::plugin::amqp1::send_queue_limit: 5000
   
              # to receive extra information about virtual memory, you must enable vmem plugin in CollectdExtraPlugins
              # collectd::plugin::vmem::verbose: true
   
              # provide name and uuid in addition to hostname for better correlation
              # to ceilometer data
              collectd::plugin::virt::hostname_format: "name uuid hostname"
   
              # to capture all extra_stats metrics, comment out below config
              collectd::plugin::virt::extra_stats: cpu_util vcpu disk
   
              # provide the human-friendly name of the virtual instance
              collectd::plugin:ConfigMap :virt::plugin_instance_format: metadata
   
              # set memcached collectd plugin to report its metrics by hostname
              # rather than host IP, ensuring metrics in the dashboard remain uniform
              collectd::plugin::memcached::instances:
                local:
                  host: "%{hiera('fqdn_canonical')}"
                  port: 11211
   
              # report root filesystem storage metrics
              collectd::plugin::df::ignoreselected: false

Procedure

1. Log in to the Red Hat OpenShift Container Platform environment where RHOSP director Operator is deployed and change to the project that hosts your RHOSP deployment:

   $ oc project openstack

2. Open the heat-env-config-deploy ConfigMap for editing:

   $ oc edit configmap heat-env-config-deploy

3. Add your stf-connectors.yaml configuration to the heat-env-config-deploy ConfigMap, appropriate to your environment, save your edits and close the file:

   stf-connectors.yaml

   apiVersion: v1
   data:
     [...]
     stf-connectors.yaml: |
       resource_registry:
         OS::TripleO::Services::Collectd: /usr/share/openstack-tripleo-heat-templates/deployment/metrics/collectd-container-puppet.yaml
   
       parameter_defaults:
           MetricsQdrConnectors:
               - host: default-interconnect-5671-service-telemetry.apps.ostest.test.metalkube.org
                 port: 443
                 role: edge
                 verifyHostname: false
                 sslProfile: sslProfile
                 saslUsername: guest@default-interconnect
                 saslPassword: <password_from_stf>
   
           MetricsQdrSSLProfiles:
               - name: sslProfile
   
           CeilometerQdrMetricsConfig:
               driver: amqp
               topic: cloud1-metering
   
           CollectdAmqpInstances:
               cloud1-telemetry:
                   format: JSON
                   presettle: false
   
           CollectdSensubilityResultsChannel: sensubility/cloud1-telemetry

   • The resource_registry configuration directly loads the collectd service because you do not include the collectd-write-qdr.yaml environment file for multiple cloud deployments.
   • Replace the host sub-parameter of MetricsQdrConnectors with the value that you retrieved in Section 4.1.3, “Retrieving the AMQ Interconnect route address”.
   • Replace the <password_from_stf> portion of the saslPassword sub-parameter of MetricsQdrConnectors with the value you retrieved in Section 4.1.2, “Retrieving the AMQ Interconnect password”.
   • Replace the caCertFileContent parameter with the contents retrieved in Section 4.1.1, “Getting CA certificate from Service Telemetry Framework for overcloud configuration”.
   • Set topic value of CeilometerQdrMetricsConfig.topic to define the topic for Ceilometer metrics. The value is a unique topic identifier for the cloud such as cloud1-metering.
   • Set CollectdAmqpInstances sub-parameter to define the topic for collectd metrics. The section name is a unique topic identifier for the cloud such as cloud1-telemetry.
   • Set CollectdSensubilityResultsChannel to define the topic for collectd-sensubility events. The value is a unique topic identifier for the cloud such as sensubility/cloud1-telemetry.

Procedure

1. Log in to the Red Hat OpenShift Container Platform environment where RHOSP director Operator is deployed and change to the project that hosts your RHOSP deployment:

   $ oc project openstack

2. Open the OpenStackConfigGenerator custom resource for editing:

   $ oc edit OpenStackConfigGenerator

3. Add the metrics/ceilometer-write-qdr.yaml and metrics/qdr-edge-only.yaml environment files as values for the heatEnvs parameter. Save your edits, and close the OpenStackConfigGenerator custom resource:

   Note

   If you already deployed a Red Hat OpenStack Platform environment using director Operator, you must delete the existing OpenStackConfigGenerator and create a new object with the full configuration in order to re-generate the OpenStackConfigVersion.

   OpenStackConfigGenerator

   apiVersion: osp-director.openstack.org/v1beta1
   kind: OpenStackConfigGenerator
   metadata:
     name: default
     namespace: openstack
   spec:
     heatEnvConfigMap: heat-env-config-deploy
     heatEnvs:
     - <existing_environment_file_references>
     - metrics/ceilometer-write-qdr.yaml
     - metrics/qdr-edge-only.yaml

4. If you already deployed a Red Hat OpenStack Platform environment using director Operator and generated a new OpenStackConfigVersion, edit the OpenStackDeploy object of your deployment, and set the value of spec.configVersion to the new OpenStackConfigVersion in order to update the overcloud deployment.