Merge pull request #282 from oldgiova/MC-7302-podmonitor

PodMonitor for monitoring traefik and probesOverrides

README.md

@@ -245,6 +245,9 @@ The following table lists the global, default, and other parameters supported by
 | `default.pdb.minAvailable` | PodDistruptionBudget minAvailable | `1` |
 | `default.imagePullSecrets` | Optional list of existing Image Pull Secrets in the format of `- name: my-custom-secret` | `[]` |
 | `default.updateStrategy` | The strategy to use to update existing pods | `rollingUpdate = { maxSurge = 1, maxUnavailable = 1 }` |
+| `default.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `default.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `default.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 | `ingress.enabled` | Optional Mender Ingress | `false` |
 | `dbmigration.enable` | Helm Chart hook that trigger a DB Migration utility just before an Helm Chart install or upgrade  | `true` |
 | `device_license_count.enabled` | Device license count feature - enterprise only | `false` |
@@ -340,6 +343,11 @@ The following table lists the parameters for the `api-gateway` component and the
 | `api_gateway.certs.existingSecret` | Preexisting secret containing the Cert (key `cert.crt`) and the Key (key `private.key`) | `nil` |
 | `api_gateway.updateStrategy` | The strategy to use to update existing pods | `nil` |
 | `api_gateway.accesslogs` | Traefik Access Logs, enabled by default | `true` |
+| `api_gateway.podMonitor.enabled` | If enabled, creates a PodMonitor resource for scraping Traefik metrics | `false` |
+| `api_gateway.podMonitor.customLabels` | PodMonitor custom labels | `nil` |
+| `api_gateway.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `api_gateway.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `api_gateway.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `3` |
 
 ### Parameters: deployments
 
@@ -394,6 +402,9 @@ The following table lists the parameters for the `deployments` component and the
 | `deployments.migrationRestartPolicy` | Migration job: restartPolicy option | `Never` |
 | `deployments.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `deployments.updateStrategy` | The strategy to use to update existing pods | `nil` |
+| `deployments.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `deployments.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `deployments.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: device-auth
 
@@ -454,6 +465,9 @@ The following table lists the parameters for the `device-auth` component and the
 | `device_auth.migrationRestartPolicy` | Migration job: restartPolicy option | `Never` |
 | `device_auth.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `device_auth.updateStrategy` | The strategy to use to update existing pods | `nil` |
+| `device_auth.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `device_auth.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `device_auth.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: gui
 
@@ -492,6 +506,9 @@ The following table lists the parameters for the `gui` component and their defau
 | `gui.containerSecurityContext.runAsUser` | User ID for the container | `65534` |
 | `gui.priorityClassName` | Optional pre-existing priorityClassName to be assigned to the resource | `nil` |
 | `gui.updateStrategy` | The strategy to use to update existing pods | `nil` |
+| `gui.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `gui.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `gui.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: inventory
 
@@ -543,6 +560,9 @@ The following table lists the parameters for the `inventory` component and their
 | `inventory.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `inventory.updateStrategy` | The strategy to use to update existing pods | `nil` |
 | `inventory.mongodbExistingSecret` | Use a different MongoDB secret for this service | `nil` |
+| `inventory.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `inventory.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `inventory.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: reporting
 
@@ -626,6 +646,9 @@ The following table lists the parameters for the `tenantadm` component and their
 | `tenantadm.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `tenantadm.migrationArgs` | Migration job: optional migration args (list). | `["migrate"]` |
 | `tenantadm.updateStrategy` | The strategy to use to update existing pods | `nil` |
+| `tenantadm.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `tenantadm.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `tenantadm.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 The default value for the rate limits are:
 
@@ -696,6 +719,9 @@ The following table lists the parameters for the `useradm` component and their d
 | `useradm.migrationRestartPolicy` | Migration job: restartPolicy option | `Never` |
 | `useradm.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `useradm.updateStrategy` | The strategy to use to update existing pods | `nil` |
+| `useradm.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `useradm.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `useradm.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: workflows
 
@@ -736,6 +762,9 @@ The following table lists the parameters for the `workflows-server` component an
 | `workflows.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `workflows.updateStrategy` | The strategy to use to update existing pods | `nil` |
 | `workflows.mountSecrets` | Optional `volumeMounts` and `volumes` to inject a credential files in the workflows service  | `nil` |
+| `workflows.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `workflows.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `workflows.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: create_artifact_worker
 
@@ -812,6 +841,9 @@ The following table lists the parameters for the `auditlogs` component and their
 | `auditlogs.migrationRestartPolicy` | Migration job: restartPolicy option | `Never` |
 | `auditlogs.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `auditlogs.updateStrategy` | The strategy to use to update existing pods | `nil` |
+| `auditlogs.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `auditlogs.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `auditlogs.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: iot-manager
 
@@ -858,6 +890,9 @@ The following table lists the parameters for the `iot-manager` component and the
 | `iot_manager.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `iot_manager.updateStrategy` | The strategy to use to update existing pods | `nil` |
 | `iot_manager.aesEncryptionKey.existingSecret` | Optional secret containing the AES encryption key. The secret key must be `AES_ENCRYPTION_KEY` | `nil` |
+| `iot_manager.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `iot_manager.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `iot_manager.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: deviceconnect
 
@@ -905,6 +940,9 @@ The following table lists the parameters for the `deviceconnect` component and t
 | `deviceconnect.migrationRestartPolicy` | Migration job: restartPolicy option | `Never` |
 | `deviceconnect.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `deviceconnect.updateStrategy` | The strategy to use to update existing pods | `nil` |
+| `deviceconnect.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `deviceconnect.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `deviceconnect.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: deviceconfig
 
@@ -950,6 +988,9 @@ The following table lists the parameters for the `deviceconfig` component and th
 | `deviceconfig.migrationRestartPolicy` | Migration job: restartPolicy option | `Never` |
 | `deviceconfig.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `deviceconfig.updateStrategy` | The strategy to use to update existing pods | `nil` |
+| `deviceconfig.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `deviceconfig.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `deviceconfig.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: devicemonitor
 
@@ -997,6 +1038,9 @@ The following table lists the parameters for the `devicemonitor` component and t
 | `devicemonitor.migrationRestartPolicy` | Migration job: restartPolicy option | `Never` |
 | `devicemonitor.migrationResources` | Migration job: optional K8s resources. If not specified, uses the deployment resources | `nil` |
 | `devicemonitor.updateStrategy` | The strategy to use to update existing pods | `nil` |
+| `devicemonitor.probesOverrides.successThreshold` | Override the `successThreshold` for every Readiness and Liveness probes. | `nil` |
+| `devicemonitor.probesOverrides.timeoutSeconds` | Override the `timeoutSeconds` for every Readiness and Liveness probes. | `nil` |
+| `devicemonitor.probesOverrides.failureThreshold` | Override the `failureThreshold` for every Readiness and Liveness probes. | `nil` |
 
 ### Parameters: generate_delta_worker
 Please notice that this feature is still under active development and it is

mender/CHANGELOG.md

@@ -7,6 +7,11 @@
 * Move from megabytes to mebibytes for consistency.
 * Added `inventory.mongodbExistingSecret` to override the default MongoDB secret.
 * Not using `HAVE_ENTERPRISE` when in hosted mode.
+* Added `podMonitor` resource for monitoring the `api-gateway` service (Traefik metrics).
+* Allow overriding fullname (thanks @ignatiusreza)
+* Removed unused `mender.name` function.
+* Added `probesOverrides` to override the default timeout for readiness and liveness probes.
+* Fix naming problem in templates using api_gateway and NodePort (thanks @j-rivero)
 
 ## Version 5.6.2
 * Upgrade to Mender version `3.7.4`.

mender/templates/_helpers.tpl

@@ -1,10 +1,4 @@
 {{/* vim: set filetype=mustache: */}}
-{{/*
-Expand the name of the chart.
-*/}}
-{{- define "mender.name" -}}
-{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }}
-{{- end }}
 
 {{/*
 Create a default fully qualified app name.
@@ -330,3 +324,14 @@ Storage Proxy Rule
 {{- define "mender.storageProxyRule" -}}
   {{- default "HostRegexp(`{domain:^artifacts.*$}`)" .Values.api_gateway.storage_proxy.customRule | quote }}
 {{- end -}}
+
+{{/*
+Use custom probes overrides
+*/}}
+{{- define "mender.probesOverrides" -}}
+{{- $_ := dict }}
+{{- $_ := (mergeOverwrite $_ .default .override) }}
+{{- if $_ }}
+{{- toYaml $_ }}
+{{- end }}
+{{- end }}

mender/templates/api-gateway/deployment.yaml

@@ -88,6 +88,9 @@ spec:
             - --providers.file.filename=/etc/traefik/config/traefik.yaml
             - --ping=true
             - --ping.manualrouting=true
+            {{- if .Values.api_gateway.podMonitor.enabled }}
+            - --entryPoints.metrics.address=:9090
+            {{- end }}
 {{- if .Values.api_gateway.extraArgs }}
 {{- .Values.api_gateway.extraArgs | toYaml | nindent 12 }}
 {{- end }}
@@ -99,24 +102,33 @@ spec:
         - containerPort: {{ .Values.api_gateway.httpsPort }}
 {{- end }}
         - containerPort: {{ .Values.api_gateway.httpPort }}
+{{- if .Values.api_gateway.podMonitor.enabled }}
+        - containerPort: 9090
+          name: prom-metrics
+          protocol: TCP
+{{- end }}
 
         # Readiness/liveness/startup probes
         livenessProbe:
           tcpSocket:
-          failureThreshold: 3
           httpGet:
             path: /healthz
             port: {{ .Values.api_gateway.httpPort }}
           initialDelaySeconds: 5
           periodSeconds: 5
+          {{- with include "mender.probesOverrides" (dict "default" .Values.default.probesOverrides "override" .Values.api_gateway.probesOverrides ) }}
+          {{- nindent 10 . }}
+          {{- end }}
         readinessProbe:
           tcpSocket:
-          failureThreshold: 1
           httpGet:
             path: /healthz
             port: {{ .Values.api_gateway.httpPort }}
           periodSeconds: 15
           initialDelaySeconds: 5
+          {{- with include "mender.probesOverrides" (dict "default" .Values.default.probesOverrides "override" .Values.api_gateway.probesOverrides ) }}
+          {{- nindent 10 . }}
+          {{- end }}
         startupProbe:
           failureThreshold: 30
           httpGet:

mender/templates/api-gateway/podMonitor.yaml

@@ -0,0 +1,15 @@
+{{- if .Values.api_gateway.podMonitor.enabled }}
+apiVersion: monitoring.coreos.com/v1
+kind: PodMonitor
+metadata:
+  name: {{ include "mender.fullname" . }}-api-gateway
+  labels:
+    {{- include "mender.labels" . | nindent 4 }}
+    {{- toYaml .Values.api_gateway.podMonitor.customLabels | nindent 4 }}
+spec:
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: {{ include "mender.fullname" . }}-api-gateway
+  podMetricsEndpoints:
+  - port: prom-metrics
+{{- end }}

mender/templates/auditlogs/_podtemplate.yaml

@@ -51,11 +51,17 @@ spec:
         path: /api/internal/v1/auditlogs/health
         port: 8080
       periodSeconds: 15
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.auditlogs.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     livenessProbe:
       httpGet:
         path: /api/internal/v1/auditlogs/alive
         port: 8080
       periodSeconds: 5
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.auditlogs.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     startupProbe:
       httpGet:
         path: /api/internal/v1/auditlogs/alive

mender/templates/deployments/_podtemplate.yaml

@@ -59,11 +59,17 @@ spec:
         path: /api/internal/v1/deployments/health
         port: 8080
       periodSeconds: 15
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.deployments.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     livenessProbe:
       httpGet:
         path: /api/internal/v1/deployments/alive
         port: 8080
       periodSeconds: 5
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.deployments.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     startupProbe:
       httpGet:
         path: /api/internal/v1/deployments/alive

mender/templates/device-auth/_podtemplate.yaml

@@ -58,11 +58,17 @@ spec:
         path: /api/internal/v1/devauth/health
         port: 8080
       periodSeconds: 15
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.device_auth.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     livenessProbe:
       httpGet:
         path: /api/internal/v1/devauth/alive
         port: 8080
       periodSeconds: 5
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.device_auth.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     startupProbe:
       httpGet:
         path: /api/internal/v1/devauth/alive

mender/templates/deviceconfig/_podtemplate.yaml

@@ -52,11 +52,17 @@ spec:
         path: /api/internal/v1/deviceconfig/health
         port: 8080
       periodSeconds: 15
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.deviceconfig.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     livenessProbe:
       httpGet:
         path: /api/internal/v1/deviceconfig/alive
         port: 8080
       periodSeconds: 5
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.deviceconfig.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     startupProbe:
       httpGet:
         path: /api/internal/v1/deviceconfig/alive

mender/templates/deviceconnect/_podtemplate.yaml

@@ -51,11 +51,17 @@ spec:
         path: /api/internal/v1/deviceconnect/health
         port: 8080
       periodSeconds: 15
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.deviceconnect.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     livenessProbe:
       httpGet:
         path: /api/internal/v1/deviceconnect/alive
         port: 8080
       periodSeconds: 5
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.deviceconnect.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     startupProbe:
       httpGet:
         path: /api/internal/v1/deviceconnect/alive

mender/templates/devicemonitor/_podtemplate.yaml

@@ -51,11 +51,17 @@ spec:
         path: /api/internal/v1/devicemonitor/health
         port: 8080
       periodSeconds: 15
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.devicemonitor.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     livenessProbe:
       httpGet:
         path: /api/internal/v1/devicemonitor/alive
         port: 8080
       periodSeconds: 5
+      {{- with include "mender.probesOverrides" (dict "default" .dot.Values.default.probesOverrides "override" .dot.Values.devicemonitor.probesOverrides ) }}
+      {{- nindent 6 . }}
+      {{- end }}
     startupProbe:
       httpGet:
         path: /api/internal/v1/devicemonitor/alive