taikun.cloud

Taikun Logo

Documentation

Metrics

Design #

This document describes the design and interaction between the custom resource definitions that the Prometheus Operator manages.

The custom resources managed by the Prometheus Operator are:

Prometheus  #

The Prometheus custom resource definition (CRD) declaratively defines a desired Prometheus setup to run in a Kubernetes cluster. It provides options to configure the number of replicas, persistent storage, and Alertmanagers to which the deployed Prometheus instances send alerts to.

For each Prometheus resource, the Operator deploys one or several StatefulSet objects in the same namespace (the number of statefulsets is equal to the number of shards but by default it is 1).

The CRD defines via label and namespace selectors which ServiceMonitorPodMonitor and Probe objects should be associated to the deployed Prometheus instances. The CRD also defines which PrometheusRules objects should be reconciled. The operator continuously reconciles the custom resources and generates one or several Secret objects holding the Prometheus configuration. A config-reloader container running as a sidecar in the Prometheus pod detects any change to the configuration and reloads Prometheus if needed.

Alertmanager  #

The Alertmanager custom resource definition (CRD) declaratively defines a desired Alertmanager setup to run in a Kubernetes cluster. It provides options to configure the number of replicas and persistent storage.

For each Alertmanager resource, the Operator deploys a StatefulSet in the same namespace. The Alertmanager pods are configured to mount a Secret called alertmanager-<alertmanager-name> which holds the Alertmanager configuration under the key alertmanager.yaml.

When there are two or more configured replicas, the Operator runs the Alertmanager instances in high-availability mode.

ThanosRuler  #

The ThanosRuler custom resource definition (CRD) declaratively defines a desired Thanos Ruler setup to run in a Kubernetes cluster. With Thanos Ruler recording and alerting rules can be processed across multiple Prometheus instances.

ThanosRuler instance requires at least one query endpoint which points to the location of Thanos Queriers or Prometheus instances.

Further information can also be found in the Thanos section.

ServiceMonitor  #

The ServiceMonitor custom resource definition (CRD) allows to declaratively define how a dynamic set of services should be monitored. Which services are selected to be monitored with the desired configuration is defined using label selections. This allows an organization to introduce conventions around how metrics are exposed, and then following these conventions new services are automatically discovered, without the need to reconfigure the system.

For Prometheus to monitor any application within Kubernetes an Endpoints object needs to exist. Endpoints objects are essentially lists of IP addresses. Typically an Endpoints object is populated by a Service object. A Service object discovers Pods by a label selector and adds those to the Endpoints object.

Service may expose one or more service ports, which are backed by a list of multiple endpoints that point to a Pod in the common case. This is reflected in the respective Endpoints object as well.

The ServiceMonitor object introduced by the Prometheus Operator in turn discovers those Endpoints objects and configures Prometheus to monitor those Pods.

The endpoints section of the ServiceMonitorSpec, is used to configure which ports of these Endpoints are going to be scraped for metrics, and with which parameters. For advanced use cases one may want to monitor ports of backing Pods, which are not directly part of the service endpoints. Therefore when specifying an endpoint in the endpoints section, they are strictly used.

Note: endpoints (lowercase) is the field in the ServiceMonitor CRD, while Endpoints (capitalized) is the Kubernetes object kind.

Both ServiceMonitors as well as discovered targets may come from any namespace. This is important to allow cross-namespace monitoring use cases, e.g. for meta-monitoring. Using the ServiceMonitorNamespaceSelector of the PrometheusSpec, one can restrict the namespaces ServiceMonitors are selected from by the respective Prometheus server. Using the namespaceSelector of the ServiceMonitorSpec, one can restrict the namespaces the Endpoints objects are allowed to be discovered from.

One can discover targets in all namespaces like this:

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: example-app
spec:
  selector:
    matchLabels:
      app: example-app
  endpoints:
  - port: web
  namespaceSelector:
    any: true

PodMonitor  #

The PodMonitor custom resource definition (CRD) allows to declaratively define how a dynamic set of pods should be monitored. Which pods are selected to be monitored with the desired configuration is defined using label selections. This allows an organization to introduce conventions around how metrics are exposed, and then following these conventions new pods are automatically discovered, without the need to reconfigure the system.

Pod is a collection of one or more containers which can expose Prometheus metrics on a number of ports.

The PodMonitor object introduced by the Prometheus Operator discovers these pods and generates the relevant configuration for the Prometheus server in order to monitor them.

The PodMetricsEndpoints section of the PodMonitorSpec, is used to configure which ports of a pod are going to be scraped for metrics, and with which parameters.

Both PodMonitors as well as discovered targets may come from any namespace. This is important to allow cross-namespace monitoring use cases, e.g. for meta-monitoring. Using the namespaceSelector of the PodMonitorSpec, one can restrict the namespaces the Pods are allowed to be discovered from.

Once can discover targets in all namespaces like this:

apiVersion: monitoring.coreos.com/v1
kind: PodMonitor
metadata:
  name: example-app
spec:
  selector:
    matchLabels:
      app: example-app
  podMetricsEndpoints:
  - port: web
  namespaceSelector:
    any: true

Probe  #

The Probe custom resource definition (CRD) allows to declarative define how groups of ingresses and static targets should be monitored. Besides the target, the Probe object requires a prober which is the service that monitors the target and provides metrics for Prometheus to scrape. Typically, this is achieved using the blackbox exporter.

PrometheusRule  #

The PrometheusRule custom resource definition (CRD) declaratively defines desired Prometheus rules to be consumed by Prometheus or Thanos Ruler instances.

Alerts and recording rules are reconciled by the Operator and dynamically loaded without requiring any restart of Prometheus/Thanos Ruler.

AlertmanagerConfig  #

The AlertmanagerConfig custom resource definition (CRD) declaratively specifies subsections of the Alertmanager configuration, allowing routing of alerts to custom receivers, and setting inhibition rules. The AlertmanagerConfig can be defined on a namespace level providing an aggregated configuration to Alertmanager. An example on how to use it is provided below. Please be aware that this CRD is not stable yet.

apiVersion: monitoring.coreos.com/v1alpha1
kind: AlertmanagerConfig
metadata:
  name: config-example
  labels:
    alertmanagerConfig: example
spec:
  route:
    groupBy: ['job']
    groupWait: 30s
    groupInterval: 5m
    repeatInterval: 12h
    receiver: 'webhook'
  receivers:
  - name: 'webhook'
    webhookConfigs:
    - url: 'http://example.com/'

PrometheusAgent  #

The PrometheusAgent custom resource definition (CRD) declaratively defines a desired Prometheus Agent setup to run in a Kubernetes cluster.

Similar to the binaries of Prometheus Server and Prometheus Agent, the Prometheus and PrometheusAgent CRs are also similar. Inspired in the Agent binary, the Agent CR has several configuration options redacted when compared with regular Prometheus CR, e.g. alerting, PrometheusRules selectors, remote-read, storage and thanos sidecars.

A more extensive read explaining why Agent support was done with a whole new CRD can be seen here.

Was this article helpful?