test-mirrors/charts/k8s-monitoring/alloyModules/modules/kubernetes/cert-manager/metrics.alloy

/*
Module: job-cert-manager
Description: Scrapes cert-manager

Note: Every argument except for "forward_to" is optional, and does have a defined default value.  However, the values for these
      arguments are not defined using the default = " ... " argument syntax, but rather using the coalesce(argument.value, " ... ").
      This is because if the argument passed in from another consuming module is set to null, the default = " ... " syntax will
      does not override the value passed in, where coalesce() will return the first non-null value.
*/
declare "kubernetes" {
  // arguments for kubernetes discovery
  argument "namespaces" {
    comment = "The namespaces to look for targets in (default: [] is all namespaces)"
    optional = true
  }

  argument "field_selectors" {
    // Docs: https://kubernetes.io/docs/concepts/overview/working-with-objects/labels/
    comment = "The label selectors to use to find matching targets (default: [])"
    optional = true
  }

  argument "label_selectors" {
    // Docs: https://kubernetes.io/docs/concepts/overview/working-with-objects/labels/
    comment = "The label selectors to use to find matching targets (default: [\"app.kubernetes.io/name=cert-manager\"])"
    optional = true
  }

  argument "port_name" {
    comment = "The of the port to scrape metrics from (default: http-metrics)"
    optional = true
  }

  // cert-manager service discovery for all of the pods
  discovery.kubernetes "cert_manager" {
    role = "pod"

    selectors {
      role = "pod"
      field = string.join(coalesce(argument.field_selectors.value, []), ",")
      label = string.join(coalesce(argument.label_selectors.value, ["app.kubernetes.io/name=cert-manager"]), ",")
    }

    namespaces {
      names = coalesce(argument.namespaces.value, [])
    }
  }

  // cert-manager relabelings (pre-scrape)
  discovery.relabel "kubernetes" {
    targets = discovery.kubernetes.cert_manager.targets

    // keep only the specified metrics port name, and pods that are Running and ready
    rule {
      source_labels = [
        "__meta_kubernetes_pod_container_port_name",
        "__meta_kubernetes_pod_phase",
        "__meta_kubernetes_pod_ready",
      ]
      separator = "@"
      regex = coalesce(argument.port_name.value, "http-metrics") + "@Running@true"
      action = "keep"
    }

    // drop any init containers
    rule {
      source_labels = ["__meta_kubernetes_pod_container_init"]
      regex = "true"
      action = "drop"
    }

    // set the namespace label
    rule {
      source_labels = ["__meta_kubernetes_namespace"]
      target_label  = "namespace"
    }

    // set the pod label
    rule {
      source_labels = ["__meta_kubernetes_pod_name"]
      target_label  = "pod"
    }

    // set the container label
    rule {
      source_labels = ["__meta_kubernetes_pod_container_name"]
      target_label  = "container"
    }

    // set a workload label
    rule {
      source_labels = [
        "__meta_kubernetes_pod_controller_kind",
        "__meta_kubernetes_pod_controller_name",
      ]
      separator = "/"
      target_label  = "workload"
    }
    // remove the hash from the ReplicaSet
    rule {
      source_labels = ["workload"]
      regex = "(ReplicaSet/.+)-.+"
      target_label  = "workload"
    }

    // set the app name if specified as metadata labels "app:" or "app.kubernetes.io/name:" or "k8s-app:"
    rule {
      action = "replace"
      source_labels = [
        "__meta_kubernetes_pod_label_app_kubernetes_io_name",
        "__meta_kubernetes_pod_label_k8s_app",
        "__meta_kubernetes_pod_label_app",
      ]
      separator = ";"
      regex = "^(?:;*)?([^;]+).*$"
      replacement = "$1"
      target_label = "app"
    }

    // set the component if specified as metadata labels "component:" or "app.kubernetes.io/component:" or "k8s-component:"
    rule {
      action = "replace"
      source_labels = [
        "__meta_kubernetes_pod_label_app_kubernetes_io_component",
        "__meta_kubernetes_pod_label_k8s_component",
        "__meta_kubernetes_pod_label_component",
      ]
      regex = "^(?:;*)?([^;]+).*$"
      replacement = "$1"
      target_label = "component"
    }

    // set a source label
    rule {
      action = "replace"
      replacement = "kubernetes"
      target_label = "source"
    }
  }

  export "output" {
    value = discovery.relabel.kubernetes.output
  }
}

declare "scrape" {
  argument "targets" {
    comment = "Must be a list() of targets"
  }

  argument "forward_to" {
    comment = "Must be a list(MetricsReceiver) where collected metrics should be forwarded to"
  }

  argument "job_label" {
    comment = "The job label to add for all cert-manager metric (default: integrations/cert-manager)"
    optional = true
  }

  argument "keep_metrics" {
    comment = "A regular expression of metrics to keep (default: see below)"
    optional = true
  }

  argument "drop_metrics" {
    comment = "A regular expression of metrics to drop (default: see below)"
    optional = true
  }

  argument "scrape_interval" {
    comment = "How often to scrape metrics from the targets (default: 60s)"
    optional = true
  }

  argument "scrape_timeout" {
    comment = "How long before a scrape times out (default: 10s)"
    optional = true
  }

  argument "max_cache_size" {
    comment = "The maximum number of elements to hold in the relabeling cache (default: 100000).  This should be at least 2x-5x your largest scrape target or samples appended rate."
    optional = true
  }

  argument "clustering" {
    // Docs: https://grafana.com/docs/agent/latest/flow/concepts/clustering/
    comment = "Whether or not clustering should be enabled (default: false)"
    optional = true
  }

  // cert-manager scrape job
  prometheus.scrape "cert_manager" {
    job_name = coalesce(argument.job_label.value, "integrations/cert-manager")
    forward_to = [prometheus.relabel.cert_manager.receiver]
    targets = argument.targets.value
    scrape_interval = coalesce(argument.scrape_interval.value, "60s")
    scrape_timeout = coalesce(argument.scrape_timeout.value, "10s")

    clustering {
      enabled = coalesce(argument.clustering.value, false)
    }
  }

  // cert-manager metric relabelings (post-scrape)
  prometheus.relabel "cert_manager" {
    forward_to = argument.forward_to.value
    max_cache_size = coalesce(argument.max_cache_size.value, 100000)

    // drop metrics that match the drop_metrics regex
    rule {
      source_labels = ["__name__"]
      regex = coalesce(argument.drop_metrics.value, "(^(go|process)_.+$)")
      action = "drop"
    }

    // keep only metrics that match the keep_metrics regex
    rule {
      source_labels = ["__name__"]
      regex = coalesce(argument.keep_metrics.value, "(up|(certmanager_(certificate_(expiration_timestamp_seconds|ready_status)|clock_time_seconds|controller_sync_call_count|http_acme_client_request_(count|duration_seconds_(count|sum)))|container_(cpu_(cfs_(periods|throttled_periods)_total|usage_seconds_total)|memory_usage_bytes|network_(receive|transmit)_bytes_total)|kube_pod_container_resource_(limits|requests)_(cpu_cores|memory_bytes)))")
      action = "keep"
    }
  }
}