Monitoring Configuration Reference

PostgreSQL Operator Monitoring Installer Configuration

When installing the PostgreSQL Operator Monitoring infrastructure you have various configuration options available, which are defined below.

General Configuration

These variables affect the general configuration of PostgreSQL Operator Monitoring.

Name Default Required Description
alertmanager_log_level info Set the log level for Alertmanager logging.
alertmanager_service_type ClusterIP Required How to expose the Alertmanager service.
alertmanager_storage_access_mode ReadWriteOnce Required Set to the access mode used by the configured storage class for Alertmanager persistent volumes.
alertmanager_storage_class_name fast Set to the name of the storage class used when creating Alertmanager persistent volumes. Omit if not using storage classes.
alertmanager_supplemental_groups 65534 Set to configure any supplemental groups that should be added to security contexts for Alertrmanager.
alertmanager_volume_size 1Gi Required Set to the size of persistent volume to create for Alertmanager.
create_rbac true Required Set to true if the installer should create the RBAC resources required to run the PostgreSQL Operator Monitoring infrastructure.
db_port 5432 Required Set to configure the PostgreSQL port used by all PostgreSQL clusters.
delete_metrics_namespace false Set to configure whether or not the metrics namespace (defined using variable metrics_namespace) is deleted when uninstalling the monitoring infrastructure.
disable_fsgroup false Set to true for deployments where you do not want to have the default PostgreSQL fsGroup (26) set. The typical usage is in OpenShift environments that have a restricted Security Context Constraints.
grafana_admin_password admin Required Set to configure the login password for the Grafana administrator.
grafana_admin_username admin Required Set to configure the login username for the Grafana administrator.
grafana_install true Required Set to true to install Grafana to visualize metrics.
grafana_service_type ClusterIP Required How to expose the Grafana service.
grafana_storage_access_mode ReadWriteOnce Required Set to the access mode used by the configured storage class for Grafana persistent volumes.
grafana_storage_class_name fast Set to the name of the storage class used when creating Grafana persistent volumes. Omit if not using storage classes.
grafana_supplemental_groups 65534 Set to configure any supplemental groups that should be added to security contexts for Grafana.
grafana_volume_size 1Gi Required Set to the size of persistent volume to create for Grafana.
metrics_image_pull_secret Name of a Secret containing credentials for container image registries.
metrics_image_pull_secret_manifest Provide a path to the image Secret manifest to be installed in the metrics namespace.
metrics_namespace 1G Required The namespace that should be created (if it doesn’t already exist) and utilized for installation of the Matrics infrastructure.
pgbadgerport 10000 Required Set to configure the port used by pgbadger in any PostgreSQL clusters.
prometheus_install false Required Set to true to install Promotheus in order to capture metrics exported from PostgreSQL clusters.
prometheus_service_type true Required How to expose the Prometheus service.
prometheus_storage_access_mode ReadWriteOnce Required Set to the access mode used by the configured storage class for Prometheus persistent volumes.
prometheus_storage_class_name fast Set to the name of the storage class used when creating Prometheus persistent volumes. Omit if not using storage classes.
prometheus_supplemental_groups 65534 Set to configure any supplemental groups that should be added to security contexts for Prometheus.
prometheus_volume_size 1Gi Required Set to the size of persistent volume to create for Prometheus.

Custom Configuration

When installing the PostgreSQL Operator Monitoring infrastructure, it is possible to further customize the various Deployments included (e.g. Alertmanager, Grafana, and/or Prometheus) using custom configuration files. Specifically, by pointing the PostgreSQL Operator Monitoring installer to one or more ConfigMaps containing any desired custom configuration settings, those settings will then be applied during configuration and installation of the PostgreSQL Operator Monitoring infrastructure.

The specific custom configuration settings available are as follows:

Name Default Required Description
alertmanager_custom_config alertmanager-config The name of a ConfigMap containing a custom alertmanager.yml configuration file.
alertmanager_custom_rules_config alertmanager-rules-config The name of a ConfigMap container custom alerting rules for Prometheus.
grafana_datasources_custom_config grafana-datasources The name of a ConfigMap containing custom Grafana datasources.
grafana_dashboards_custom_config grafana-dashboards The name of a ConfigMap containing custom Grafana dashboards.
prometheus_custom_configmap crunchy-prometheus The name of a ConfigMap containing a custom prometheus.yml configuration file.

Please note that when using custom ConfigMaps per the above configuration settings, any defaults for the specific configuration being customized are no longer applied.

Using Alertmanager

The Alertmanager deployment requires a custom configuration file to configure reciever integrations that are supported by Prometheus Alertmanager. The installer will create a configmap containing an example Alertmanager configuration file created by the pgMonitor project, this file can be found in the pgMonitor repository. This example file, along with the Alertmanager configuration docs, will help you to configure alerting for you specific use cases.

Alertmanager cannot be installed without also deploying the Crunchy Prometheus deployment. Once both are deployed, Prometheus is automatically configured to send alerts to the Alertmanager.

Using RedHat Certified Containers & Custom Images

By default, the PostgreSQL Operator Monitoring installer will deploy the official Grafana and Prometheus containers that are publically available on dockerhub:

However, if RedHat certified containers are needed, the following certified images have also been verified with the PostgreSQL Operator Metric infrastructure, and can therefore be utilized instead:

The following configuration settings can be applied to properly configure the image prefix, name and tag as needed to use the RedHat certified containers:

Name Default Required Description
alertmanager_image_prefix prom Required Configure the image prefix to use for the Alertmanager container.
alertmanager_image_name alertmanager Required Configure the image name to use for the Alertmanager container.
alertmanager_image_tag v0.21.0 Required Configures the image tag to use for the Alertmanager container.
grafana_image_prefix grafana Required Configures the image prefix to use for the Grafana container.
grafana_image_name grafana Required Configures the image name to use for the Grafana container.
grafana_image_tag 6.7.4 Required Configures the image tag to use for the Grafana container.
prometheus_image_prefix prom Required Configures the image prefix to use for the Prometheus container.
prometheus_image_name promtheus Required Configures the image name to use for the Prometheus container.
prometheus_image_tag v2.20.0 Required Configures the image tag to use for the Prometheus container.

Additionally, these same settings can be utilized as needed to support custom image names, tags, and additional container registries.

Helm Only Configuration Settings

When using Helm, the following settings can be defined to control the image prefix and image tag utilized for the pgo-deployer container that is run to install, update or uninstall the PostgreSQL Operator Monitoring infrastructure:

Name Default Required Description
pgo_image_prefix registry.developers.crunchydata.com/crunchydata Required Configures the image prefix used by the pgo-deployer container
pgo_image_tag centos7-4.5.0 Required Configures the image tag used by the pgo-deployer container