Move Prometheus as an Add-On

charts/add-ons/prometheus/.helmignore

@@ -0,0 +1,22 @@
+# Patterns to ignore when building packages.
+# This supports shell glob matching, relative path matching, and
+# negation (prefixed with !). Only one pattern per line.
+.DS_Store
+# Common VCS dirs
+.git/
+.gitignore
+.bzr/
+.bzrignore
+.hg/
+.hgignore
+.svn/
+# Common backup files
+*.swp
+*.bak
+*.tmp
+*~
+# Various IDEs
+.project
+.idea/
+*.tmproj
+.vscode/

charts/add-ons/prometheus/Chart.yaml

@@ -0,0 +1,9 @@
+apiVersion: v1
+appVersion: "1.0"
+description: A Helm chart for the prometheus add-on in Linkerd
+name: prometheus
+version: 0.1.0
+maintainers:
+  - name: Linkerd authors
+    email: cncf-linkerd-dev@lists.cncf.io
+    url: https://linkerd.io/

charts/add-ons/prometheus/requirements.yaml

@@ -0,0 +1,4 @@
+dependencies:
+  - name: partials
+    version: 0.1.0
+    repository: file://../../partials

charts/linkerd2/templates/prometheus.yaml → charts/add-ons/prometheus/templates/prometheus.yaml

@@ -15,15 +15,10 @@ metadata:
     {{.Values.global.createdByAnnotation}}: {{default (printf "linkerd/helm %s" .Values.global.linkerdVersion) .Values.global.cliVersion}}
 data:
   prometheus.yml: |-
-    {{- if .Values.prometheusAlertmanagers }}
-    alerting:
-      alertmanagers:
-        {{- toYaml .Values.prometheusAlertmanagers | trim | nindent 8 }}
-    {{- end }}
     global:
-      scrape_interval: 10s
-      scrape_timeout: 10s
-      evaluation_interval: 10s
+      {{- if .Values.globalConfig -}}
+      {{- toYaml .Values.globalConfig | trim | nindent 6 }}
+      {{- end}}
 
     rule_files:
     - /etc/prometheus/*_rules.yml
@@ -34,7 +29,6 @@ data:
       static_configs:
       - targets: ['localhost:9090']
 
-    {{ if .Values.grafana.enabled -}} 
     - job_name: 'grafana'
       kubernetes_sd_configs:
       - role: pod
@@ -45,7 +39,6 @@ data:
         - __meta_kubernetes_pod_container_name
         action: keep
         regex: ^grafana$
-      {{- end}}
 
     #  Required for: https://grafana.com/grafana/dashboards/315
     - job_name: 'kubernetes-nodes-cadvisor'
@@ -54,7 +47,6 @@ data:
         ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
         insecure_skip_verify: true
       bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token
-
       kubernetes_sd_configs:
       - role: node
       relabel_configs:
@@ -153,6 +145,27 @@ data:
       # Copy tmp labels into real labels
       - action: labelmap
         regex: __tmp_pod_label_(.+)
+
+    {{- if .Values.scrapeConfigs }}
+    {{- toYaml .Values.scrapeConfigs | trim | nindent 4 }}
+    {{- end }}
+
+    {{-  if (or .Values.alertManagers .Values.alertRelabelConfigs) }}
+    alerting:
+      alert_relabel_configs:
+        {{- if .Values.alertRelabelConfigs }}
+        {{- toYaml .Values.alertRelabelConfigs | trim | nindent 6 }}
+        {{- end }}
+      alertmanagers:
+        {{- if .Values.alertManagers }}
+        {{- toYaml .Values.alertManagers | trim | nindent 6 }}
+        {{- end }}
+    {{- end }}
+
+    {{- if .Values.remoteWrite }}
+    remote_write:
+      {{- toYaml .Values.remoteWrite | trim | nindent 4 }}
+    {{- end }}
 ---
 kind: Service
 apiVersion: v1
@@ -210,14 +223,10 @@ spec:
       {{- include "linkerd.node-selector" . | nindent 6 }}
       containers:
       - args:
-        - --storage.tsdb.path=/data
-        - --storage.tsdb.retention.time=6h
-        - --config.file=/etc/prometheus/prometheus.yml
-        - --log.level={{lower .Values.prometheusLogLevel}}
-        {{- range $key, $value := .Values.prometheusExtraArgs}}
+        {{- range $key, $value := .Values.args}}
         - --{{ $key }}{{ if $value }}={{ $value }}{{ end }}
         {{- end }}
-        image: {{.Values.prometheusImage}}
+        image: {{ default "prom/prometheus:v2.15.2" .Values.image }}
         imagePullPolicy: {{.Values.global.imagePullPolicy}}
         livenessProbe:
           httpGet:
@@ -235,13 +244,13 @@ spec:
             port: 9090
           initialDelaySeconds: 30
           timeoutSeconds: 30
-        {{- if .Values.prometheusResources -}}
-        {{- include "partials.resources" .Values.prometheusResources | nindent 8 }}
+        {{- if .Values.resources -}}
+        {{- include "partials.resources" .Values.resources | nindent 8 }}
         {{- end }}
         securityContext:
           runAsUser: 65534
         volumeMounts:
-      {{- range .Values.prometheusRuleConfigMapMounts }}
+      {{- range .Values.ruleConfigMapMounts }}
         - name: {{ .name }}
           mountPath: /etc/prometheus/{{ .subPath }}
           subPath: {{ .subPath }}
@@ -260,7 +269,7 @@ spec:
       {{ end -}}
       serviceAccountName: linkerd-prometheus
       volumes:
-    {{- range .Values.prometheusRuleConfigMapMounts }}
+    {{- range .Values.ruleConfigMapMounts }}
       - name: {{ .name }}
         configMap:
           name: {{ .configMap }}

charts/add-ons/prometheus/values.yaml

@@ -0,0 +1,10 @@
+image: prom/prometheus:v2.15.2
+args:
+  storage.tsdb.path: /data
+  storage.tsdb.retention.time: 6h
+  config.file: /etc/prometheus/prometheus.yml
+  log.level: info
+globalConfig:
+  scrape_interval: 10s
+  scrape_timeout: 10s
+  evaluation_interval: 10s

charts/linkerd2/README.md

@@ -148,11 +148,6 @@ their default values.
 | `identity.issuer.tls.keyPEM`                | Key for the issuer certificate (ECDSA). It must be provided during install.                                                                                                           |                                      |
 | `installNamespace`                          | Set to false when installing Linkerd in a custom namespace. See the [Linkerd documentation](https://linkerd.io/2/tasks/install-helm/#customizing-the-namespace) for more information. | `true`                               |
 | `omitWebhookSideEffects`                    | Omit the `sideEffects` flag in the webhook manifests                                                                                                                                  | `false`                              |
-| `prometheusAlertmanagers`                   | Alertmanager instances the Prometheus server sends alerts to configured via the static_configs parameter.                                                                             | `[]`                                 |
-| `prometheusExtraArgs`                       | Extra command line options for Prometheus                                                                                                                                             | `{}`                                 |
-| `prometheusImage`                           | Docker image for the Prometheus container                                                                                                                                             | `prom/prometheus:v2.15.2`            |
-| `prometheusLogLevel`                        | Log level for Prometheus                                                                                                                                                              | `info`                               |
-| `prometheusRuleConfigMapMounts`             | Alerting/recording rule ConfigMap mounts (sub-path names must end in `_rules.yml` or `_rules.yaml`)                                                                                   | `[]`                                 |
 | `proxyInjector.externalSecret`              | Do not create a secret resource for the profileValidator webhook. If this is set to `true`, the value `proxyInjector.caBundle` must be set (see below).                                                 | false                              |
 | `proxyInjector.crtPEM`                      | Certificate for the proxy injector. If not provided then Helm will generate one.                                                                                                      |                                      |
 | `proxyInjector.keyPEM`                      | Certificate key for the proxy injector. If not provided then Helm will generate one.                                                                                                      |                                      |
@@ -191,6 +186,28 @@ The following table lists the configurable parameters for the Grafana Add-On.
 | `grafana.resources.memory.limit`    | Maximum amount of memory that grafana container can use                                                                                                                        ||
 | `grafana.resources.memory.request`  | Amount of memory that the grafana container requests                                                                                                                               ||
 
+### Prometheus Add-On
+
+The following table lists the configurable parameters for the Prometheus Add-On.
+
+| Parameter                             | Description                                                                                                                                                                           | Default                              |
+|:--------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------|
+| `prometheus.enabled`                     | Flag to enable prometheus instance to be installed                                                                                                                                                | `true`                             |
+| `prometheus.alert_relabel_configs`                   | Alert relabeling is applied to alerts before they are sent to the Alertmanager.                                                                            | `[]`                                 |
+| `prometheus.alertManagers`                   | Alertmanager instances the Prometheus server sends alerts to configured via the static_configs parameter.                                                                             | `[]`                                 |
+| `prometheus.args`                       |  Command line options for Prometheus binary                                                                                                                                             | `storage.tsdb.path: /data, storage.tsdb.retention.time: 6h, config.file: /etc/prometheus/prometheus.yml, log.level: *controller_log_level`                                 |
+| `prometheus.globalConfig`             | The global configuration specifies parameters that are valid in all other configuration contexts.                                                                 | `scrape_interval: 10s, scrape_timeout: 10s, evaluation_interval: 10s`                                 |
+| `prometheus.image`                | Docker image for the prometheus instance                                                                                                                                                 | `prom/prometheus:v2.15.2`                             |
+| `prometheus.resources.cpu.limit`       | Maximum amount of CPU units that the prometheus container can use                                                                                                                     ||
+| `prometheus.resources.cpu.request`     | Amount of CPU units that the prometheus container requests                                                                                                                            ||
+| `prometheus.resources.memory.limit`    | Maximum amount of memory that prometheus container can use                                                                                                                        ||
+| `prometheus.resources.memory.request`  | Amount of memory that the prometheus container requests                                                                                                                               ||
+| `prometheus.ruleConfigMapMounts`             | Alerting/recording rule ConfigMap mounts (sub-path names must end in `_rules.yml` or `_rules.yaml`)                                                                                   | `[]`                                 |
+| `prometheus.scrapeConfigs`             | A scrape_config section specifies a set of targets and parameters describing how to scrape them.                                                        | `[]`                                 |
+
+Most of the above configuration match directly with the official Prometheus
+configuration which can be found [here](https://prometheus.io/docs/prometheus/latest/configuration/configuration)
+
 ### Tracing Add-On
 
 The following table lists the configurable parameters for the Tracing Add-On.

charts/linkerd2/requirements.lock

@@ -2,11 +2,14 @@ dependencies:
 - name: partials
   repository: file://../partials
   version: 0.1.0
+- name: prometheus
+  repository: file://../add-ons/prometheus
+  version: 0.1.0
 - name: grafana
   repository: file://../add-ons/grafana
   version: 0.1.0
 - name: tracing
   repository: file://../add-ons/tracing
   version: 0.1.0
-digest: sha256:f92907b6d243e3b57b4288603ba76eced7c2f4ef913e76505c314971bb4afa21
-generated: "2020-05-11T14:13:54.306010536-05:00"
+digest: sha256:d2428770ae7d5134c5af6521c78a4c5f95da4c75f21bdea0f74fad6ab6e2e044
+generated: "2020-06-24T11:07:53.924602129Z"

charts/linkerd2/requirements.yaml

@@ -2,6 +2,10 @@ dependencies:
 - name: partials
   version: 0.1.0
   repository: file://../partials
+- name: prometheus
+  version: 0.1.0
+  repository: file://../add-ons/prometheus
+  condition: prometheus.enabled
 - name: grafana
   version: 0.1.0
   repository: file://../add-ons/grafana

charts/linkerd2/templates/controller.yaml

@@ -72,7 +72,7 @@ spec:
         - -prometheus-url=http://linkerd-prometheus.{{.Values.global.namespace}}.svc.{{.Values.global.clusterDomain}}:9090
         - -destination-addr=linkerd-dst.{{.Values.global.namespace}}.svc.{{.Values.global.clusterDomain}}:8086
         - -controller-namespace={{.Values.global.namespace}}
-        - -log-level={{.Values.controllerLogLevel}}
+        - -log-level={{.Values.global.controllerLogLevel}}
         {{- include "partials.linkerd.trace" . | nindent 8 -}}
         image: {{.Values.controllerImage}}:{{default .Values.global.linkerdVersion .Values.global.controllerImageVersion}}
         imagePullPolicy: {{.Values.global.imagePullPolicy}}

charts/linkerd2/templates/destination.yaml

@@ -72,7 +72,7 @@ spec:
         - -addr=:8086
         - -controller-namespace={{.Values.global.namespace}}
         - -enable-h2-upgrade={{.Values.enableH2Upgrade}}
-        - -log-level={{.Values.controllerLogLevel}}
+        - -log-level={{.Values.global.controllerLogLevel}}
         {{- include "partials.linkerd.trace" . | nindent 8 -}}
         image: {{.Values.controllerImage}}:{{default .Values.global.linkerdVersion .Values.global.controllerImageVersion}}
         imagePullPolicy: {{.Values.global.imagePullPolicy}}

charts/linkerd2/templates/heartbeat.yaml

@@ -41,7 +41,7 @@ spec:
             - "heartbeat"
             - "-prometheus-url=http://linkerd-prometheus.{{.Values.global.namespace}}.svc.{{.Values.global.clusterDomain}}:9090"
             - "-controller-namespace={{.Values.global.namespace}}"
-            - "-log-level={{.Values.controllerLogLevel}}"
+            - "-log-level={{.Values.global.controllerLogLevel}}"
             {{- if .Values.heartbeatResources -}}
             {{- include "partials.resources" .Values.heartbeatResources | nindent 12 }}
             {{- end }}

charts/linkerd2/templates/identity.yaml

@@ -89,7 +89,7 @@ spec:
       containers:
       - args:
         - identity
-        - -log-level={{.Values.controllerLogLevel}}
+        - -log-level={{.Values.global.controllerLogLevel}}
         {{- include "partials.linkerd.trace" . | nindent 8 -}}
         image: {{.Values.controllerImage}}:{{default .Values.global.linkerdVersion .Values.global.controllerImageVersion}}
         imagePullPolicy: {{.Values.global.imagePullPolicy}}

charts/linkerd2/templates/linkerd-config-addons.yaml

@@ -24,5 +24,7 @@ data:
       grafanaUrl: "{{.Values.global.grafanaUrl}}"
     grafana:
       {{- include "linkerd.addons.sanitize-config" .Values.grafana}}
+    prometheus:
+      {{- include "linkerd.addons.sanitize-config" .Values.prometheus}}
     tracing:
       {{- include "linkerd.addons.sanitize-config" .Values.tracing}}

charts/linkerd2/templates/proxy-injector.yaml

@@ -51,7 +51,7 @@ spec:
       containers:
       - args:
         - proxy-injector
-        - -log-level={{.Values.controllerLogLevel}}
+        - -log-level={{.Values.global.controllerLogLevel}}
         image: {{.Values.controllerImage}}:{{default .Values.global.linkerdVersion .Values.global.controllerImageVersion}}
         imagePullPolicy: {{.Values.global.imagePullPolicy}}
         livenessProbe:

charts/linkerd2/templates/sp-validator.yaml

@@ -70,7 +70,7 @@ spec:
       containers:
       - args:
         - sp-validator
-        - -log-level={{.Values.controllerLogLevel}}
+        - -log-level={{.Values.global.controllerLogLevel}}
         image: {{.Values.controllerImage}}:{{default .Values.global.linkerdVersion .Values.global.controllerImageVersion}}
         imagePullPolicy: {{.Values.global.imagePullPolicy}}
         livenessProbe:

charts/linkerd2/templates/tap.yaml

@@ -76,7 +76,7 @@ spec:
       - args:
         - tap
         - -controller-namespace={{.Values.global.namespace}}
-        - -log-level={{.Values.controllerLogLevel}}
+        - -log-level={{.Values.global.controllerLogLevel}}
         {{- include "partials.linkerd.trace" . | nindent 8 -}}
         image: {{.Values.controllerImage}}:{{default .Values.global.linkerdVersion .Values.global.controllerImageVersion}}
         imagePullPolicy: {{.Values.global.imagePullPolicy}}

charts/linkerd2/templates/web.yaml

@@ -72,7 +72,7 @@ spec:
         - -jaeger-addr={{.Values.tracing.jaeger.name}}.{{.Values.global.namespace}}.svc.{{.Values.global.clusterDomain}}:16686
         {{- end}}
         - -controller-namespace={{.Values.global.namespace}}
-        - -log-level={{.Values.controllerLogLevel}}
+        - -log-level={{.Values.global.controllerLogLevel}}
         {{- if .Values.enforcedHostRegexp }}
         - -enforced-host={{.Values.enforcedHostRegexp}}
         {{- else -}}

charts/linkerd2/values-ha.yaml

@@ -46,13 +46,14 @@ grafana:
 heartbeatResources: *controller_resources
 
 # prometheus configuration
-prometheusResources:
-  cpu:
-    limit: "4"
-    request: 300m
-  memory:
-    limit: 8192Mi
-    request: 300Mi
+prometheus:
+  resources:
+    cpu:
+      limit: "4"
+      request: 300m
+    memory:
+      limit: 8192Mi
+      request: 300Mi
 
 # proxy injector configuration
 proxyInjectorResources: *controller_resources