1. Documentation
×

Configuring the monitoring stack

Prerequisites

  • The monitoring stack imposes additional resource requirements. Consult the computing resources recommendations in Scaling the Cluster Monitoring Operator and verify that you have sufficient resources.

Maintenance and support for monitoring

The supported way of configuring OKD Monitoring is by configuring it using the options described in this document. Do not use other configurations, as they are unsupported. Configuration paradigms might change across Prometheus releases, and such cases can only be handled gracefully if all configuration possibilities are controlled. If you use configurations other than those described in this section, your changes will disappear because the cluster-monitoring-operator reconciles any differences. The Operator resets everything to the defined state by default and by design.

Support considerations for monitoring

The following modifications are explicitly not supported:

  • Creating additional ServiceMonitor, PodMonitor, and PrometheusRule objects in the openshift-* and kube-* projects.

  • Modifying any resources or objects deployed in the openshift-monitoring or openshift-user-workload-monitoring projects. The resources created by the OKD monitoring stack are not meant to be used by any other resources, as there are no guarantees about their backward compatibility.

    The Alertmanager configuration is deployed as a secret resource in the openshift-monitoring project. To configure additional routes for Alertmanager, you need to decode, modify, and then encode that secret. This procedure is a supported exception to the preceding statement.

  • Modifying resources of the stack. The OKD monitoring stack ensures its resources are always in the state it expects them to be. If they are modified, the stack will reset them.

  • Deploying user-defined workloads to openshift-*, and kube-* projects. These projects are reserved for Red Hat provided components and they should not be used for user-defined workloads.

  • Modifying the monitoring stack Grafana instance.

  • Installing custom Prometheus instances on OKD. A custom instance is a Prometheus custom resource (CR) managed by the Prometheus Operator.

  • Enabling symptom based monitoring by using the Probe custom resource definition (CRD) in Prometheus Operator.

  • Modifying Alertmanager configurations by using the AlertmanagerConfig CRD in Prometheus Operator.

Backward compatibility for metrics, recording rules, or alerting rules is not guaranteed.

Support policy for monitoring Operators

Monitoring Operators ensure that OKD monitoring resources function as designed and tested. If Cluster Version Operator (CVO) control of an Operator is overridden, the Operator does not respond to configuration changes, reconcile the intended state of cluster objects, or receive updates.

While overriding CVO control for an Operator can be helpful during debugging, this is unsupported and the cluster administrator assumes full control of the individual component configurations and upgrades.

Overriding the Cluster Version Operator

The spec.overrides parameter can be added to the configuration for the CVO to allow administrators to provide a list of overrides to the behavior of the CVO for a component. Setting the spec.overrides[].unmanaged parameter to true for a component blocks cluster upgrades and alerts the administrator after a CVO override has been set:

Disabling ownership via cluster version overrides prevents upgrades. Please remove overrides before continuing.

Setting a CVO override puts the entire cluster in an unsupported state and prevents the monitoring stack from being reconciled to its intended state. This impacts the reliability features built into Operators and prevents updates from being received. Reported issues must be reproduced after removing any overrides for support to proceed.

Preparing to configure the monitoring stack

You can configure the monitoring stack by creating and updating monitoring config maps.

Creating a cluster monitoring config map

To configure core OKD monitoring components, you must create the cluster-monitoring-config ConfigMap object in the openshift-monitoring project.

When you save your changes to the cluster-monitoring-config ConfigMap object, some or all of the pods in the openshift-monitoring project might be redeployed. It can sometimes take a while for these components to redeploy.
Prerequisites

  • You have access to the cluster as a user with the cluster-admin role.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Check whether the cluster-monitoring-config ConfigMap object exists:

    $ oc -n openshift-monitoring get configmap cluster-monitoring-config

  2. If the ConfigMap object does not exist:

    1. Create the following YAML manifest. In this example the file is called cluster-monitoring-config.yaml:

      apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |

    2. Apply the configuration to create the ConfigMap object:

      $ oc apply -f cluster-monitoring-config.yaml

Creating a user-defined workload monitoring config map

To configure the components that monitor user-defined projects, you must create the user-workload-monitoring-config ConfigMap object in the openshift-user-workload-monitoring project.

When you save your changes to the user-workload-monitoring-config ConfigMap object, some or all of the pods in the openshift-user-workload-monitoring project might be redeployed. It can sometimes take a while for these components to redeploy. You can create and configure the config map before you first enable monitoring for user-defined projects, to prevent having to redeploy the pods often.
Prerequisites

  • You have access to the cluster as a user with the cluster-admin role.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Check whether the user-workload-monitoring-config ConfigMap object exists:

    $ oc -n openshift-user-workload-monitoring get configmap user-workload-monitoring-config

  2. If the user-workload-monitoring-config ConfigMap object does not exist:

    1. Create the following YAML manifest. In this example the file is called user-workload-monitoring-config.yaml:

      apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |

    2. Apply the configuration to create the ConfigMap object:

      $ oc apply -f user-workload-monitoring-config.yaml

      Configurations applied to the user-workload-monitoring-config ConfigMap object are not activated unless a cluster administrator has enabled monitoring for user-defined projects.
Additional resources

Configuring the monitoring stack

In OKD 4.7, you can configure the monitoring stack using the cluster-monitoring-config or user-workload-monitoring-config ConfigMap objects. Config maps configure the Cluster Monitoring Operator (CMO), which in turn configures the components of the stack.

Prerequisites

  • If you are configuring core OKD monitoring components:

    • You have access to the cluster as a user with the cluster-admin role.

    • You have created the cluster-monitoring-config ConfigMap object.

  • If you are configuring components that monitor user-defined projects:

    • You have access to the cluster as a user with the cluster-admin role, or as a user with the user-workload-monitoring-config-edit role in the openshift-user-workload-monitoring project.

    • You have created the user-workload-monitoring-config ConfigMap object.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Edit the ConfigMap object.

    • To configure core OKD monitoring components:

      1. Edit the cluster-monitoring-config ConfigMap object in the openshift-monitoring project:

        $ oc -n openshift-monitoring edit configmap cluster-monitoring-config

      2. Add your configuration under data/config.yaml as a key-value pair <component_name>: <component_configuration>:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    <component>:
      <configuration_for_the_component>

        Substitute <component> and <configuration_for_the_component> accordingly.

        The following example ConfigMap object configures a persistent volume claim (PVC) for Prometheus. This relates to the Prometheus instance that monitors core OKD components only:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    prometheusK8s: (1)
      volumeClaimTemplate:
       spec:
         storageClassName: fast
         volumeMode: Filesystem
         resources:
           requests:
             storage: 40Gi
        1 Defines the Prometheus component and the subsequent lines define its configuration.

    • To configure components that monitor user-defined projects:

      1. Edit the user-workload-monitoring-config ConfigMap object in the openshift-user-workload-monitoring project:

        $ oc -n openshift-user-workload-monitoring edit configmap user-workload-monitoring-config

      2. Add your configuration under data/config.yaml as a key-value pair <component_name>: <component_configuration>:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    <component>:
      <configuration_for_the_component>

        Substitute <component> and <configuration_for_the_component> accordingly.

        The following example ConfigMap object configures a data retention period and minimum container resource requests for Prometheus. This relates to the Prometheus instance that monitors user-defined projects only:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    prometheus: (1)
      retention: 24h (2)
      resources:
        requests:
          cpu: 200m (3)
          memory: 2Gi (4)
        1 Defines the Prometheus component and the subsequent lines define its configuration.
        2 Configures a twenty-four hour data retention period for the Prometheus instance that monitors user-defined projects.
        3 Defines a minimum resource request of 200 millicores for the Prometheus container.
        4 Defines a minimum pod resource request of 2 GiB of memory for the Prometheus container.

        The Prometheus config map component is called prometheusK8s in the cluster-monitoring-config ConfigMap object and prometheus in the user-workload-monitoring-config ConfigMap object.

  2. Save the file to apply the changes to the ConfigMap object. The pods affected by the new configuration are restarted automatically.

    Configurations applied to the user-workload-monitoring-config ConfigMap object are not activated unless a cluster administrator has enabled monitoring for user-defined projects.

    When changes are saved to a monitoring config map, the pods and other resources in the related project might be redeployed. The running monitoring processes in that project might also be restarted.
Additional resources

Configurable monitoring components

This table shows the monitoring components you can configure and the keys used to specify the components in the cluster-monitoring-config and user-workload-monitoring-config ConfigMap objects:

Table 1. Configurable monitoring components
Component cluster-monitoring-config config map key user-workload-monitoring-config config map key

Prometheus Operator

prometheusOperator

prometheusOperator

Prometheus

prometheusK8s

prometheus

Alertmanager

alertmanagerMain

kube-state-metrics

kubeStateMetrics

openshift-state-metrics

openshiftStateMetrics

Grafana

grafana

Telemeter Client

telemeterClient

Prometheus Adapter

k8sPrometheusAdapter

Thanos Querier

thanosQuerier

Thanos Ruler

thanosRuler

The Prometheus key is called prometheusK8s in the cluster-monitoring-config ConfigMap object and prometheus in the user-workload-monitoring-config ConfigMap object.

Moving monitoring components to different nodes

You can move any of the monitoring stack components to specific nodes.

Prerequisites

  • If you are configuring core OKD monitoring components:

    • You have access to the cluster as a user with the cluster-admin role.

    • You have created the cluster-monitoring-config ConfigMap object.

  • If you are configuring components that monitor user-defined projects:

    • You have access to the cluster as a user with the cluster-admin role, or as a user with the user-workload-monitoring-config-edit role in the openshift-user-workload-monitoring project.

    • You have created the user-workload-monitoring-config ConfigMap object.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Edit the ConfigMap object:

    • To move a component that monitors core OKD projects:

      1. Edit the cluster-monitoring-config ConfigMap object in the openshift-monitoring project:

        $ oc -n openshift-monitoring edit configmap cluster-monitoring-config

      2. Specify the nodeSelector constraint for the component under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    <component>:
      nodeSelector:
        <node_key>: <node_value>
        <node_key>: <node_value>
        <...>

        Substitute <component> accordingly and substitute <node_key>: <node_value> with the map of key-value pairs that specifies a group of destination nodes. Often, only a single key-value pair is used.

        The component can only run on nodes that have each of the specified key-value pairs as labels. The nodes can have additional labels as well.

        Many of the monitoring components are deployed by using multiple pods across different nodes in the cluster to maintain high availability. When moving monitoring components to labeled nodes, ensure that enough matching nodes are available to maintain resilience for the component. If only one label is specified, ensure that enough nodes contain that label to distribute all of the pods for the component across separate nodes. Alternatively, you can specify multiple labels each relating to individual nodes.

        If monitoring components remain in a Pending state after configuring the nodeSelector constraint, check the pod logs for errors relating to taints and tolerations.

        For example, to move monitoring components for core OKD projects to specific nodes that are labeled nodename: controlplane1, nodename: worker1, nodename: worker2, and nodename: worker2, use:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    prometheusOperator:
      nodeSelector:
        nodename: controlplane1
    prometheusK8s:
      nodeSelector:
        nodename: worker1
        nodename: worker2
    alertmanagerMain:
      nodeSelector:
        nodename: worker1
        nodename: worker2
    kubeStateMetrics:
      nodeSelector:
        nodename: worker1
    grafana:
      nodeSelector:
        nodename: worker1
    telemeterClient:
      nodeSelector:
        nodename: worker1
    k8sPrometheusAdapter:
      nodeSelector:
        nodename: worker1
        nodename: worker2
    openshiftStateMetrics:
      nodeSelector:
        nodename: worker1
    thanosQuerier:
      nodeSelector:
        nodename: worker1
        nodename: worker2

    • To move a component that monitors user-defined projects:

      1. Edit the user-workload-monitoring-config ConfigMap object in the openshift-user-workload-monitoring project:

        $ oc -n openshift-user-workload-monitoring edit configmap user-workload-monitoring-config

      2. Specify the nodeSelector constraint for the component under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    <component>:
      nodeSelector:
        <node_key>: <node_value>
        <node_key>: <node_value>
        <...>

        Substitute <component> accordingly and substitute <node_key>: <node_value> with the map of key-value pairs that specifies the destination nodes. Often, only a single key-value pair is used.

        The component can only run on nodes that have each of the specified key-value pairs as labels. The nodes can have additional labels as well.

        Many of the monitoring components are deployed by using multiple pods across different nodes in the cluster to maintain high availability. When moving monitoring components to labeled nodes, ensure that enough matching nodes are available to maintain resilience for the component. If only one label is specified, ensure that enough nodes contain that label to distribute all of the pods for the component across separate nodes. Alternatively, you can specify multiple labels each relating to individual nodes.

        If monitoring components remain in a Pending state after configuring the nodeSelector constraint, check the pod logs for errors relating to taints and tolerations.

        For example, to move monitoring components for user-defined projects to specific worker nodes labeled nodename: worker1, nodename: worker2, and nodename: worker2, use:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    prometheusOperator:
      nodeSelector:
        nodename: worker1
    prometheus:
      nodeSelector:
        nodename: worker1
        nodename: worker2
    thanosRuler:
      nodeSelector:
        nodename: worker1
        nodename: worker2

  2. Save the file to apply the changes. The components affected by the new configuration are moved to the new nodes automatically.

    Configurations applied to the user-workload-monitoring-config ConfigMap object are not activated unless a cluster administrator has enabled monitoring for user-defined projects.

    When changes are saved to a monitoring config map, the pods and other resources in the related project might be redeployed. The running monitoring processes in that project might also be restarted.
Additional resources

Assigning tolerations to monitoring components

You can assign tolerations to any of the monitoring stack components to enable moving them to tainted nodes.

Prerequisites

  • If you are configuring core OKD monitoring components:

    • You have access to the cluster as a user with the cluster-admin role.

    • You have created the cluster-monitoring-config ConfigMap object.

  • If you are configuring components that monitor user-defined projects:

    • You have access to the cluster as a user with the cluster-admin role, or as a user with the user-workload-monitoring-config-edit role in the openshift-user-workload-monitoring project.

    • You have created the user-workload-monitoring-config ConfigMap object.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Edit the ConfigMap object:

    • To assign tolerations to a component that monitors core OKD projects:

      1. Edit the cluster-monitoring-config ConfigMap object in the openshift-monitoring project:

        $ oc -n openshift-monitoring edit configmap cluster-monitoring-config

      2. Specify tolerations for the component:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    <component>:
      tolerations:
        <toleration_specification>

        Substitute <component> and <toleration_specification> accordingly.

        For example, oc adm taint nodes node1 key1=value1:NoSchedule adds a taint to node1 with the key key1 and the value value1. This prevents monitoring components from deploying pods on node1 unless a toleration is configured for that taint. The following example configures the alertmanagerMain component to tolerate the example taint:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    alertmanagerMain:
      tolerations:
      - key: "key1"
        operator: "Equal"
        value: "value1"
        effect: "NoSchedule"

    • To assign tolerations to a component that monitors user-defined projects:

      1. Edit the user-workload-monitoring-config ConfigMap object in the openshift-user-workload-monitoring project:

        $ oc -n openshift-user-workload-monitoring edit configmap user-workload-monitoring-config

      2. Specify tolerations for the component:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    <component>:
      tolerations:
        <toleration_specification>

        Substitute <component> and <toleration_specification> accordingly.

        For example, oc adm taint nodes node1 key1=value1:NoSchedule adds a taint to node1 with the key key1 and the value value1. This prevents monitoring components from deploying pods on node1 unless a toleration is configured for that taint. The following example configures the thanosRuler component to tolerate the example taint:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    thanosRuler:
      tolerations:
      - key: "key1"
        operator: "Equal"
        value: "value1"
        effect: "NoSchedule"

  2. Save the file to apply the changes. The new component placement configuration is applied automatically.

    Configurations applied to the user-workload-monitoring-config ConfigMap object are not activated unless a cluster administrator has enabled monitoring for user-defined projects.

    When changes are saved to a monitoring config map, the pods and other resources in the related project might be redeployed. The running monitoring processes in that project might also be restarted.
Additional resources

Configuring persistent storage

Running cluster monitoring with persistent storage means that your metrics are stored to a persistent volume (PV) and can survive a pod being restarted or recreated. This is ideal if you require your metrics or alerting data to be guarded from data loss. For production environments, it is highly recommended to configure persistent storage. Because of the high IO demands, it is advantageous to use local storage.

If you are running cluster monitoring with an attached PVC for Prometheus, you might experience OOM kills during cluster upgrade. When persistent storage is in use for Prometheus, Prometheus memory usage doubles during cluster upgrade and for several hours after upgrade is complete. To avoid the OOM kill issue, allow worker nodes with double the size of memory that was available prior to the upgrade. For example, if you are running monitoring on the minimum recommended nodes, which is 2 cores with 8 GB of RAM, increase memory to 16 GB. For more information, see BZ#1925061.

See Recommended configurable storage technology.

Persistent storage prerequisites

  • Dedicate sufficient local persistent storage to ensure that the disk does not become full. How much storage you need depends on the number of pods. For information on system requirements for persistent storage, see Prometheus database storage requirements.

  • Make sure you have a persistent volume (PV) ready to be claimed by the persistent volume claim (PVC), one PV for each replica. Because Prometheus has two replicas and Alertmanager has three replicas, you need five PVs to support the entire monitoring stack. The PVs should be available from the Local Storage Operator. This does not apply if you enable dynamically provisioned storage.

  • Use the block type of storage.

  • Configure local persistent storage.

    If you use a local volume for persistent storage, do not use a raw block volume, which is described with volumeMode: block in the LocalVolume object. Prometheus cannot use raw block volumes.

Configuring a local persistent volume claim

For monitoring components to use a persistent volume (PV), you must configure a persistent volume claim (PVC).

Prerequisites

  • If you are configuring core OKD monitoring components:

    • You have access to the cluster as a user with the cluster-admin role.

    • You have created the cluster-monitoring-config ConfigMap object.

  • If you are configuring components that monitor user-defined projects:

    • You have access to the cluster as a user with the cluster-admin role, or as a user with the user-workload-monitoring-config-edit role in the openshift-user-workload-monitoring project.

    • You have created the user-workload-monitoring-config ConfigMap object.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Edit the ConfigMap object:

    • To configure a PVC for a component that monitors core OKD projects:

      1. Edit the cluster-monitoring-config ConfigMap object in the openshift-monitoring project:

        $ oc -n openshift-monitoring edit configmap cluster-monitoring-config

      2. Add your PVC configuration for the component under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    <component>:
      volumeClaimTemplate:
        spec:
          storageClassName: <storage_class>
          resources:
            requests:
              storage: <amount_of_storage>

        See the Kubernetes documentation on PersistentVolumeClaims for information on how to specify volumeClaimTemplate.

        The following example configures a PVC that claims local persistent storage for the Prometheus instance that monitors core OKD components:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    prometheusK8s:
      volumeClaimTemplate:
        spec:
          storageClassName: local-storage
          resources:
            requests:
              storage: 40Gi

        In the above example, the storage class created by the Local Storage Operator is called local-storage.

        The following example configures a PVC that claims local persistent storage for Alertmanager:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    alertmanagerMain:
      volumeClaimTemplate:
        spec:
          storageClassName: local-storage
          resources:
            requests:
              storage: 10Gi

    • To configure a PVC for a component that monitors user-defined projects:

      1. Edit the user-workload-monitoring-config ConfigMap object in the openshift-user-workload-monitoring project:

        $ oc -n openshift-user-workload-monitoring edit configmap user-workload-monitoring-config

      2. Add your PVC configuration for the component under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    <component>:
      volumeClaimTemplate:
        spec:
          storageClassName: <storage_class>
          resources:
            requests:
              storage: <amount_of_storage>

        See the Kubernetes documentation on PersistentVolumeClaims for information on how to specify volumeClaimTemplate.

        The following example configures a PVC that claims local persistent storage for the Prometheus instance that monitors user-defined projects:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    prometheus:
      volumeClaimTemplate:
        spec:
          storageClassName: local-storage
          resources:
            requests:
              storage: 40Gi

        In the above example, the storage class created by the Local Storage Operator is called local-storage.

        The following example configures a PVC that claims local persistent storage for Thanos Ruler:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    thanosRuler:
      volumeClaimTemplate:
        spec:
          storageClassName: local-storage
          resources:
            requests:
              storage: 10Gi

        Storage requirements for the thanosRuler component depend on the number of rules that are evaluated and how many samples each rule generates.

  2. Save the file to apply the changes. The pods affected by the new configuration are restarted automatically and the new storage configuration is applied.

    Configurations applied to the user-workload-monitoring-config ConfigMap object are not activated unless a cluster administrator has enabled monitoring for user-defined projects.

    When changes are saved to a monitoring config map, the pods and other resources in the related project might be redeployed. The running monitoring processes in that project might also be restarted.

Modifying the retention time for Prometheus metrics data

By default, the OKD monitoring stack configures the retention time for Prometheus data to be 15 days. You can modify the retention time to change how soon the data is deleted.

Prerequisites

  • If you are configuring core OKD monitoring components:

    • You have access to the cluster as a user with the cluster-admin role.

    • You have created the cluster-monitoring-config ConfigMap object.

  • If you are configuring components that monitor user-defined projects:

    • You have access to the cluster as a user with the cluster-admin role, or as a user with the user-workload-monitoring-config-edit role in the openshift-user-workload-monitoring project.

    • You have created the user-workload-monitoring-config ConfigMap object.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Edit the ConfigMap object:

    • To modify the retention time for the Prometheus instance that monitors core OKD projects:

      1. Edit the cluster-monitoring-config ConfigMap object in the openshift-monitoring project:

        $ oc -n openshift-monitoring edit configmap cluster-monitoring-config

      2. Add your retention time configuration under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    prometheusK8s:
      retention: <time_specification>

        Substitute <time_specification> with a number directly followed by ms (milliseconds), s (seconds), m (minutes), h (hours), d (days), w (weeks), or y (years).

        The following example sets the retention time to 24 hours for the Prometheus instance that monitors core OKD components:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    prometheusK8s:
      retention: 24h

    • To modify the retention time for the Prometheus instance that monitors user-defined projects:

      1. Edit the user-workload-monitoring-config ConfigMap object in the openshift-user-workload-monitoring project:

        $ oc -n openshift-user-workload-monitoring edit configmap user-workload-monitoring-config

      2. Add your retention time configuration under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    prometheus:
      retention: <time_specification>

        Substitute <time_specification> with a number directly followed by ms (milliseconds), s (seconds), m (minutes), h (hours), d (days), w (weeks), or y (years).

        The following example sets the retention time to 24 hours for the Prometheus instance that monitors user-defined projects:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    prometheus:
      retention: 24h

  2. Save the file to apply the changes. The pods affected by the new configuration are restarted automatically.

    Configurations applied to the user-workload-monitoring-config ConfigMap object are not activated unless a cluster administrator has enabled monitoring for user-defined projects.

    When changes are saved to a monitoring config map, the pods and other resources in the related project might be redeployed. The running monitoring processes in that project might also be restarted.
Additional resources

Controlling the impact of unbound metrics attributes in user-defined projects

Developers can create labels to define attributes for metrics in the form of key-value pairs. The number of potential key-value pairs corresponds to the number of possible values for an attribute. An attribute that has an unlimited number of potential values is called an unbound attribute. For example, a customer_id attribute is unbound because it has an infinite number of possible values.

Every assigned key-value pair has a unique time series. The use of many unbound attributes in labels can result in an exponential increase in the number of time series created. This can impact Prometheus performance and can consume a lot of disk space.

Cluster administrators can use the following measures to control the impact of unbound metrics attributes in user-defined projects:

  • Limit the number of samples that can be accepted per target scrape in user-defined projects

  • Create alerts that fire when a scrape sample threshold is reached or when the target cannot be scraped

Limiting scrape samples can help prevent the issues caused by adding many unbound attributes to labels. Developers can also prevent the underlying cause by limiting the number of unbound attributes that they define for metrics. Using attributes that are bound to a limited set of possible values reduces the number of potential key-value pair combinations.

Setting a scrape sample limit for user-defined projects

You can limit the number of samples that can be accepted per target scrape in user-defined projects.

If you set a sample limit, no further sample data is ingested for that target scrape after the limit is reached.
Prerequisites

  • You have access to the cluster as a user with the cluster-admin role, or as a user with the user-workload-monitoring-config-edit role in the openshift-user-workload-monitoring project.

  • You have created the user-workload-monitoring-config ConfigMap object.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Edit the user-workload-monitoring-config ConfigMap object in the openshift-user-workload-monitoring project:

    $ oc -n openshift-user-workload-monitoring edit configmap user-workload-monitoring-config

  2. Add the enforcedSampleLimit configuration to data/config.yaml to limit the number of samples that can be accepted per target scrape in user-defined projects:

    apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    prometheus:
      enforcedSampleLimit: 50000 (1)
    1 A value is required if this parameter is specified. This enforcedSampleLimit example limits the number of samples that can be accepted per target scrape in user-defined projects to 50,000.

  3. Save the file to apply the changes. The limit is applied automatically.

    Configurations applied to the user-workload-monitoring-config ConfigMap object are not activated unless a cluster administrator has enabled monitoring for user-defined projects.

    When changes are saved to the user-workload-monitoring-config ConfigMap object, the pods and other resources in the openshift-user-workload-monitoring project might be redeployed. The running monitoring processes in that project might also be restarted.

Creating scrape sample alerts

You can create alerts that notify you when:

  • The target cannot be scraped or is not available for the specified for duration

  • A scrape sample threshold is reached or is exceeded for the specified for duration

Prerequisites

  • You have access to the cluster as a user with the cluster-admin role, or as a user with the user-workload-monitoring-config-edit role in the openshift-user-workload-monitoring project.

  • You have enabled monitoring for user-defined projects.

  • You have created the user-workload-monitoring-config ConfigMap object.

  • You have limited the number of samples that can be accepted per target scrape in user-defined projects, by using enforcedSampleLimit.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Create a YAML file with alerts that inform you when the targets are down and when the enforced sample limit is approaching. The file in this example is called monitoring-stack-alerts.yaml:

    apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  labels:
    prometheus: k8s
    role: alert-rules
  name: monitoring-stack-alerts (1)
  namespace: ns1 (2)
spec:
  groups:
  - name: general.rules
    rules:
    - alert: TargetDown (3)
      annotations:
        message: '{{ printf "%.4g" $value }}% of the {{ $labels.job }}/{{ $labels.service
          }} targets in {{ $labels.namespace }} namespace are down.' (4)
      expr: 100 * (count(up == 0) BY (job, namespace, service) / count(up) BY (job,
        namespace, service)) > 10
      for: 10m (5)
      labels:
        severity: warning (6)
    - alert: ApproachingEnforcedSamplesLimit (7)
      annotations:
        message: '{{ $labels.container }} container of the {{ $labels.pod }} pod in the {{ $labels.namespace }} namespace consumes {{ $value | humanizePercentage }} of the samples limit budget.' (8)
      expr: scrape_samples_scraped/50000 > 0.8 (9)
      for: 10m (10)
      labels:
        severity: warning (11)
    1 Defines the name of the alerting rule.
    2 Specifies the user-defined project where the alerting rule will be deployed.
    3 The TargetDown alert will fire if the target cannot be scraped or is not available for the for duration.
    4 The message that will be output when the TargetDown alert fires.
    5 The conditions for the TargetDown alert must be true for this duration before the alert is fired.
    6 Defines the severity for the TargetDown alert.
    7 The ApproachingEnforcedSamplesLimit alert will fire when the defined scrape sample threshold is reached or exceeded for the specified for duration.
    8 The message that will be output when the ApproachingEnforcedSamplesLimit alert fires.
    9 The threshold for the ApproachingEnforcedSamplesLimit alert. In this example the alert will fire when the number of samples per target scrape has exceeded 80% of the enforced sample limit of 50000. The for duration must also have passed before the alert will fire. The <number> in the expression scrape_samples_scraped/<number> > <threshold> must match the enforcedSampleLimit value defined in the user-workload-monitoring-config ConfigMap object.
    10 The conditions for the ApproachingEnforcedSamplesLimit alert must be true for this duration before the alert is fired.
    11 Defines the severity for the ApproachingEnforcedSamplesLimit alert.

  2. Apply the configuration to the user-defined project:

    $ oc apply -f monitoring-stack-alerts.yaml
Additional resources

Attaching additional labels to your time series and alerts

Using the external labels feature of Prometheus, you can attach custom labels to all time series and alerts leaving Prometheus.

Prerequisites

  • If you are configuring core OKD monitoring components:

    • You have access to the cluster as a user with the cluster-admin role.

    • You have created the cluster-monitoring-config ConfigMap object.

  • If you are configuring components that monitor user-defined projects:

    • You have access to the cluster as a user with the cluster-admin role, or as a user with the user-workload-monitoring-config-edit role in the openshift-user-workload-monitoring project.

    • You have created the user-workload-monitoring-config ConfigMap object.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Edit the ConfigMap object:

    • To attach custom labels to all time series and alerts leaving the Prometheus instance that monitors core OKD projects:

      1. Edit the cluster-monitoring-config ConfigMap object in the openshift-monitoring project:

        $ oc -n openshift-monitoring edit configmap cluster-monitoring-config

      2. Define a map of labels you want to add for every metric under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    prometheusK8s:
      externalLabels:
        <key>: <value> (1)
        1 Substitute <key>: <value> with a map of key-value pairs where <key> is a unique name for the new label and <value> is its value.

        Do not use prometheus or prometheus_replica as key names, because they are reserved and will be overwritten.

        For example, to add metadata about the region and environment to all time series and alerts, use:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    prometheusK8s:
      externalLabels:
        region: eu
        environment: prod

    • To attach custom labels to all time series and alerts leaving the Prometheus instance that monitors user-defined projects:

      1. Edit the user-workload-monitoring-config ConfigMap object in the openshift-user-workload-monitoring project:

        $ oc -n openshift-user-workload-monitoring edit configmap user-workload-monitoring-config

      2. Define a map of labels you want to add for every metric under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    prometheus:
      externalLabels:
        <key>: <value> (1)
        1 Substitute <key>: <value> with a map of key-value pairs where <key> is a unique name for the new label and <value> is its value.

        Do not use prometheus or prometheus_replica as key names, because they are reserved and will be overwritten.

        In the openshift-user-workload-monitoring project, Prometheus handles metrics and Thanos Ruler handles alerting and recording rules. Setting externalLabels for prometheus in the user-workload-monitoring-config ConfigMap object will only configure external labels for metrics and not for any rules.

        For example, to add metadata about the region and environment to all time series and alerts related to user-defined projects, use:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    prometheus:
      externalLabels:
        region: eu
        environment: prod

  2. Save the file to apply the changes. The new configuration is applied automatically.

    Configurations applied to the user-workload-monitoring-config ConfigMap object are not activated unless a cluster administrator has enabled monitoring for user-defined projects.

    When changes are saved to a monitoring config map, the pods and other resources in the related project might be redeployed. The running monitoring processes in that project might also be restarted.
Additional resources

Setting log levels for monitoring components

You can configure the log level for Prometheus Operator, Prometheus, Thanos Querier, and Thanos Ruler.

You cannot use this procedure to configure the log level for the Alertmanager component.

The following log levels can be applied to each of those components in the cluster-monitoring-config and user-workload-monitoring-config ConfigMap objects:

  • debug. Log debug, informational, warning, and error messages.

  • info. Log informational, warning, and error messages.

  • warn. Log warning and error messages only.

  • error. Log error messages only.

The default log level is info.

Prerequisites

  • If you are setting a log level for Prometheus Operator, Prometheus, or Thanos Querier in the openshift-monitoring project:

    • You have access to the cluster as a user with the cluster-admin role.

    • You have created the cluster-monitoring-config ConfigMap object.

  • If you are setting a log level for Prometheus Operator, Prometheus, or Thanos Ruler in the openshift-user-workload-monitoring project:

    • You have access to the cluster as a user with the cluster-admin role, or as a user with the user-workload-monitoring-config-edit role in the openshift-user-workload-monitoring project.

    • You have created the user-workload-monitoring-config ConfigMap object.

  • You have installed the OpenShift CLI (oc).

Procedure

  1. Edit the ConfigMap object:

    • To set a log level for a component in the openshift-monitoring project:

      1. Edit the cluster-monitoring-config ConfigMap object in the openshift-monitoring project:

        $ oc -n openshift-monitoring edit configmap cluster-monitoring-config

      2. Add logLevel: <log_level> for a component under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    <component>: (1)
      logLevel: <log_level> (2)
        1 The monitoring component that you are applying a log level to.
        2 The log level to apply to the component.

    • To set a log level for a component in the openshift-user-workload-monitoring project:

      1. Edit the user-workload-monitoring-config ConfigMap object in the openshift-user-workload-monitoring project:

        $ oc -n openshift-user-workload-monitoring edit configmap user-workload-monitoring-config

      2. Add logLevel: <log_level> for a component under data/config.yaml:

        apiVersion: v1
kind: ConfigMap
metadata:
  name: user-workload-monitoring-config
  namespace: openshift-user-workload-monitoring
data:
  config.yaml: |
    <component>: (1)
      logLevel: <log_level> (2)
        1 The monitoring component that you are applying a log level to.
        2 The log level to apply to the component.

  2. Save the file to apply the changes. The pods for the component restarts automatically when you apply the log-level change.

    Configurations applied to the user-workload-monitoring-config ConfigMap object are not activated unless a cluster administrator has enabled monitoring for user-defined projects.

    When changes are saved to a monitoring config map, the pods and other resources in the related project might be redeployed. The running monitoring processes in that project might also be restarted.

  3. Confirm that the log-level has been applied by reviewing the deployment or pod configuration in the related project. The following example checks the log level in the prometheus-operator deployment in the openshift-user-workload-monitoring project:

    $ oc -n openshift-user-workload-monitoring get deploy prometheus-operator -o yaml |  grep "log-level"
    Example output
            - --log-level=debug

  4. Check that the pods for the component are running. The following example lists the status of pods in the openshift-user-workload-monitoring project:

    $ oc -n openshift-user-workload-monitoring get pods

    If an unrecognized loglevel value is included in the ConfigMap object, the pods for the component might not restart successfully.
Additional resources

Next steps