Important: This documentation is about an older version. It's relevant only to the release noted, many of the features and functions have been updated or replaced. Please view the current version.

Documentation Grafana Loki documentation Installation Install using Helm Monitoring Configure monitoring and alerting

Configure monitoring and alerting

By default this Helm Chart configures meta-monitoring of metrics (service monitoring) and logs (self monitoring). This topic will walk you through configuring monitoring using a monitoring solution local to the same cluster where Loki is installed.

The ServiceMonitor resource works with either the Prometheus Operator or the Grafana Agent Operator, and defines how Loki’s metrics should be scraped. Scraping this Loki cluster using the scrape config defined in the SerivceMonitor resource is required for the included dashboards to work. A MetricsInstance can be configured to write the metrics to a remote Prometheus instance such as Grafana Cloud Metrics.

Self monitoring is enabled by default. This will deploy a GrafanaAgent, LogsInstance, and PodLogs resource which will instruct the Grafana Agent Operator (installed seperately) on how to scrape this Loki cluster’s logs and send them back to itself. Scraping this Loki cluster using the scrape config defined in the PodLogs resource is required for the included dashboards to work.

Rules and alerts are automatically deployed.

Before you begin:

• Helm 3 or above. See Installing Helm.
• A running Kubernetes cluster with a running Loki deployment.
• A running Grafana instance.
• A running Prometheus Operator installed using the kube-prometheus-stack Helm chart.

Prometheus Operator Prequisites

The dashboards require certain metric labels to display Kubernetes metrics. The best way to accomplish this is to install the kube-prometheus-stack Helm chart with the following values file, replacing CLUSTER_NAME with the name of your cluster. The cluster name is what you specify during the helm installation, so a cluster installed with the command helm install loki-cluster grafana/loki would be called loki-cluster.

YAML Copy

kubelet:
  serviceMonitor:
    cAdvisorRelabelings:
      - action: replace
        replacement: <CLUSTER_NAME>
        targetLabel: cluster
      - targetLabel: metrics_path
        sourceLabels:
          - "__metrics_path__"
      - targetLabel: "instance"
        sourceLabels:
          - "node"

defaultRules:
  additionalRuleLabels:
    cluster: <CLUSTER_NAME>

"kube-state-metrics":
  prometheus:
    monitor:
      relabelings:
        - action: replace
          replacement: <CLUSTER_NAME>
          targetLabel: cluster
        - targetLabel: "instance"
          sourceLabels:
            - "__meta_kubernetes_pod_node_name"

"prometheus-node-exporter":
  prometheus:
    monitor:
      relabelings:
        - action: replace
          replacement: <CLUSTER_NAME>
          targetLabel: cluster
        - targetLabel: "instance"
          sourceLabels:
            - "__meta_kubernetes_pod_node_name"

prometheus:
  monitor:
    relabelings:
      - action: replace
        replacement: <CLUSTER_NAME>
        targetLabel: cluster

The kube-prometheus-stack installs ServicMonitor and PrometheusRule resources for monitoring Kubernetes, and it depends on the kube-state-metrics and prometheus-node-exporter helm charts which also install ServiceMonitor resources for collecting kubelet and node-exporter metrics. The above values file adds the necessary additional labels required for these metrics to work with the included dashboards.

If you are using this helm chart in an environment which does not allow for the installation of kube-prometheus-stack or custom CRDs, you should run helm template on the kube-prometheus-stack helm chart with the above values file, and review all generated ServiceMonitor and PrometheusRule resources. These resources may have to be modified with the correct ports and selectors to find the various services such as kubelet and node-exporter in your environment.

To install the dashboards:

1. Dashboards are enabled by default. Set monitoring.dashboards.namespace to the namespace of the Grafana instance if it is in a different namespace than this Loki cluster.

2. Dashbards must be mounted to your Grafana container. The dashboards are in ConfigMaps named loki-dashboards-1 and loki-dashboards-2 for Loki, and enterprise-logs-dashboards-1 and enterprise-logs-dashboards-2 for GEL. Mount them to /var/lib/grafana/dashboards/loki-1 and /var/lib/grafana/dashboards/loki-2 in your Grafana container.

3. Create a dashboard provisioning file called dashboards.yaml in /etc/grafana/provisioning/dashboards of your Grafana container with the following contents (note: you may need to edit the orgId):

   YAML Copy

   ---
   apiVersion: 1
   providers:
     - disableDeletion: true
       editable: false
       folder: Loki
       name: loki-1
       options:
         path: /var/lib/grafana/dashboards/loki-1
       orgId: 1
       type: file
     - disableDeletion: true
       editable: false
       folder: Loki
       name: loki-2
       options:
         path: /var/lib/grafana/dashboards/loki-2
       orgId: 1
       type: file

To add add additional Prometheus rules:

1. Modify the configuration file values.yaml:

   YAML Copy

   monitoring:
     rules:
       additionalGroups:
         - name: loki-rules
           rules:
             - record: job:loki_request_duration_seconds_bucket:sum_rate
               expr: sum(rate(loki_request_duration_seconds_bucket[1m])) by (le, job)
             - record: job_route:loki_request_duration_seconds_bucket:sum_rate
               expr: sum(rate(loki_request_duration_seconds_bucket[1m])) by (le, job, route)
             - record: node_namespace_pod_container:container_cpu_usage_seconds_total:sum_rate
               expr: sum(rate(container_cpu_usage_seconds_total[1m])) by (node, namespace, pod, container)

To disable monitoring:

1. Modify the configuration file values.yaml:

   YAML Copy

   selfMonitoring:
     enabled: false
   
   serviceMonitor:
     enabled: false

To use a remote Prometheus and Loki instance such as Grafana Cloud

1. Create a secrets.yaml file with credentials to access the Grafana Cloud services:

   YAML Copy

   ---
   apiVersion: v1
   kind: Secret
   metadata:
     name: primary-credentials-metrics
     namespace: default
   stringData:
     username: "<instance ID>"
     password: "<API key>"
   ---
   apiVersion: v1
   kind: Secret
   metadata:
     name: primary-credentials-logs
     namespace: default
   stringData:
     username: "<instance ID>"
     password: "<API key>"

2. Add the secret to Kubernetes with kubectl create -f secret.yaml.

3. Add a remoteWrite section to serviceMonitor in values.yaml:

   YAML Copy

   monitoring:
   ...
     serviceMonitor:
       enabled: true
       ...
       metricsInstance:
         remoteWrite:
         - url: <metrics remote write endpoint>
           basicAuth:
             username:
               name: primary-credentials-metrics
               key: username
             password:
               name: primary-credentials-metrics
               key: password

4. Add a client to monitoring.selfMonitoring.logsInstance.clients:

   YAML Copy

   monitoring:
   ---
   selfMonitoring:
     enabled: true
     logsInstance:
       clients:
         - url: <logs remote write endpoint>
           basicAuth:
             username:
               name: primary-credentials-logs
               key: username
             password:
               name: primary-credentials-logs
               key: password
   lokiCanary:
     enabled: false

5. Install the Loki meta-motoring connection on Grafana Cloud.