[datadog] Show error notices about port discrepancies (#64)

* [datadog] Show error notices about port discrepancies - Show notices for APM and Cluster Agent. - Remove explicitly set default port values from the liveness and readiness probe values to avoid possible misconfiguration. - Use mergeOverwrite to merge default health ports from `datadog.apm.port` and `clusterAgent.healthPort` * Flag node agent port discrepancy as well
DataDog · Oct 21, 2020 · 3015738 · 3015738
1 parent 87cd763
commit 3015738
Show file tree

Hide file tree

Showing 8 changed files with 80 additions and 31 deletions.
diff --git a/charts/datadog/CHANGELOG.md b/charts/datadog/CHANGELOG.md
@@ -1,5 +1,9 @@
 # Datadog changelog
 
+## 2.4.27
+
+* Remove port defaults from liveness/readiness probes and show error notices on misconfiguration if user overrides are supplying custom node settings.
+
 ## 2.4.26
 
 * Revert to Helm2 hash in `requirements.yaml` to retain compatibility with Helm 2

diff --git a/charts/datadog/Chart.yaml b/charts/datadog/Chart.yaml
@@ -1,6 +1,6 @@
 apiVersion: v1
 name: datadog
-version: 2.4.26
+version: 2.4.27
 appVersion: "7"
 description: Datadog Agent
 keywords:

diff --git a/charts/datadog/README.md b/charts/datadog/README.md
@@ -1,6 +1,6 @@
 # Datadog
 
-![Version: 2.4.26](https://img.shields.io/badge/Version-2.4.26-informational?style=flat-square) ![AppVersion: 7](https://img.shields.io/badge/AppVersion-7-informational?style=flat-square)
+![Version: 2.4.27](https://img.shields.io/badge/Version-2.4.27-informational?style=flat-square) ![AppVersion: 7](https://img.shields.io/badge/AppVersion-7-informational?style=flat-square)
 
 [Datadog](https://www.datadoghq.com/) is a hosted infrastructure monitoring platform. This chart adds the Datadog Agent to all nodes in your cluster via a DaemonSet. It also optionally depends on the [kube-state-metrics chart](https://github.com/kubernetes/charts/tree/master/stable/kube-state-metrics). For more information about monitoring Kubernetes with Datadog, please refer to the [Datadog documentation website](https://docs.datadoghq.com/agent/basic_agent_usage/kubernetes/).
 
@@ -310,6 +310,7 @@ helm install --name <RELEASE_NAME> \
 |-----|------|---------|-------------|
 | agents.affinity | object | `{}` | Allow the DaemonSet to schedule using affinity rules |
 | agents.containers.agent.env | list | `[]` | Additional environment variables for the agent container |
+| agents.containers.agent.healthPort | int | `5555` | Port number to use in the node agent for the healthz endpoint |
 | agents.containers.agent.livenessProbe | object | Every 15s / 6 KO / 1 OK | Override default agent liveness probe settings |
 | agents.containers.agent.logLevel | string | `nil` | Set logging verbosity, valid log levels are: trace, debug, info, warn, error, critical, and off |
 | agents.containers.agent.readinessProbe | object | Every 15s / 6 KO / 1 OK | Override default agent readiness probe settings |
@@ -368,12 +369,12 @@ helm install --name <RELEASE_NAME> \
 | clusterAgent.dnsConfig | object | `{}` | Specify dns configuration options for datadog cluster agent containers e.g ndots |
 | clusterAgent.enabled | bool | `false` | Set this to true to enable Datadog Cluster Agent |
 | clusterAgent.env | list | `[]` | Set environment variables specific to Cluster Agent |
-| clusterAgent.healthPort | int | `5555` | Port number use the cluster-agent to server healthz endpoint |
+| clusterAgent.healthPort | int | `5555` | Port number to use in the Cluster Agent for the healthz endpoint |
 | clusterAgent.image.pullPolicy | string | `"IfNotPresent"` | Cluster Agent image pullPolicy |
 | clusterAgent.image.pullSecrets | list | `[]` | Cluster Agent repository pullSecret (ex: specify docker registry credentials) |
 | clusterAgent.image.repository | string | `"datadog/cluster-agent"` | Cluster Agent image repository to use |
 | clusterAgent.image.tag | string | `"1.9.0"` | Cluster Agent image tag to use |
-| clusterAgent.livenessProbe | object | Every 15s / 6 KO / 1 OK | Override default agent liveness probe settings |
+| clusterAgent.livenessProbe | object | Every 15s / 6 KO / 1 OK | Override default Cluster Agent liveness probe settings |
 | clusterAgent.metricsProvider.aggregator | string | `"avg"` | Define the aggregator the cluster agent will use to process the metrics. The options are (avg, min, max, sum) |
 | clusterAgent.metricsProvider.createReaderRbac | bool | `true` | Create `external-metrics-reader` RBAC automatically (to allow HPA to read data from Cluster Agent) |
 | clusterAgent.metricsProvider.enabled | bool | `false` | Set this to true to enable Metrics Provider |
@@ -387,7 +388,7 @@ helm install --name <RELEASE_NAME> \
 | clusterAgent.priorityClassName | string | `nil` | Name of the priorityClass to apply to the Cluster Agent |
 | clusterAgent.rbac.create | bool | `true` | If true, create & use RBAC resources |
 | clusterAgent.rbac.serviceAccountName | string | `"default"` | Specify service account name to use (usually pre-existing, created if create is true) |
-| clusterAgent.readinessProbe | object | Every 15s / 6 KO / 1 OK | Override default cluster-agent readiness probe settings |
+| clusterAgent.readinessProbe | object | Every 15s / 6 KO / 1 OK | Override default Cluster Agent readiness probe settings |
 | clusterAgent.replicas | int | `1` | Specify the of cluster agent replicas, if > 1 it allow the cluster agent to work in HA mode. |
 | clusterAgent.resources | object | `{}` | Datadog cluster-agent resource requests and limits. |
 | clusterAgent.strategy | object | `{"rollingUpdate":{"maxSurge":1,"maxUnavailable":0},"type":"RollingUpdate"}` | Allow the Cluster Agent deployment to perform a rolling update on helm update |

diff --git a/charts/datadog/templates/NOTES.txt b/charts/datadog/templates/NOTES.txt
@@ -34,6 +34,27 @@ Then run:
         --set datadog.apiKey=YOUR-KEY-HERE stable/datadog
 {{- end }}
 
+{{- $healthPort := .Values.agents.containers.agent.healthPort }}
+{{- with $liveness := .Values.agents.containers.agent.livenessProbe.httpGet }}
+{{- if and $liveness.port (ne $healthPort $liveness.port) }}
+
+##############################################################################
+####               ERROR: Node Agent liveness probe misconfiguration      ####
+##############################################################################
+
+Node Agent liveness probe port ({{ $liveness.port }}) is different from the configured health port ({{ $healthPort }}).
+{{- end }}
+{{- end }}
+{{- with $readiness := .Values.agents.containers.agent.readinessProbe.httpGet }}
+{{- if and $readiness.port (ne $healthPort $readiness.port) }}
+
+##############################################################################
+####               ERROR: Node Agent readiness probe misconfiguration     ####
+##############################################################################
+
+Node Agent readiness probe port ({{ $readiness.port }}) is different from the configured health port ({{ $healthPort }}).
+{{- end }}
+{{- end }}
 {{- if .Values.clusterAgent.enabled }}
 
   {{- if .Values.clusterAgent.metricsProvider.enabled }}
@@ -51,11 +72,42 @@ This deployment will be incomplete until you get your APP key from Datadog.
 Create an application key at https://app.datadoghq.com/account/settings#api
     {{- end }}
   {{- end }}
+  {{- $healthPort := .Values.clusterAgent.healthPort }}
+  {{- with $liveness := .Values.clusterAgent.livenessProbe.httpGet }}
+  {{- if and $liveness.port (ne $healthPort $liveness.port) }}
 
-{{- end }}
+##############################################################################
+####               ERROR: Cluster Agent liveness probe misconfiguration   ####
+##############################################################################
 
+Cluster Agent liveness probe port ({{ $liveness.port }}) is different from the configured health port ({{ $healthPort }}).
+  {{- end }}
+  {{- end }}
+  {{- with $readiness := .Values.clusterAgent.readinessProbe.httpGet }}
+  {{- if and $readiness.port (ne $healthPort $readiness.port) }}
+
+##############################################################################
+####               ERROR: Cluster Agent readiness probe misconfiguration  ####
+##############################################################################
+
+Cluster Agent readiness probe port ({{ $readiness.port }}) is different from the configured health port ({{ $healthPort }}).
+  {{- end }}
+  {{- end }}
+{{- end }}
 {{- if .Values.datadog.apm.enabled }}
-The Datadog Agent is listening on port {{ .Values.datadog.apm.port }} for APM service.
+  {{- $apmPort := .Values.datadog.apm.port }}
+  {{- with $liveness := .Values.agents.containers.traceAgent.livenessProbe.tcpSocket }}
+  {{- if and $liveness.port (ne $apmPort $liveness.port) }}
+
+##############################################################################
+####               ERROR: Trace Agent liveness probe misconfiguration     ####
+##############################################################################
+
+Trace Agent liveness probe port ({{ $liveness.port }}) is different from the configured APM port ({{ $apmPort }}).
+  {{- end }}
+  {{- end }}
+
+The Datadog Agent is listening on port {{ $apmPort }} for APM service.
 {{- end }}
 
 {{- if .Values.datadog.autoconf }}

diff --git a/charts/datadog/templates/cluster-agent-deployment.yaml b/charts/datadog/templates/cluster-agent-deployment.yaml
@@ -199,9 +199,11 @@ spec:
 {{ toYaml .Values.clusterAgent.env | indent 10 }}
 {{- end }}
         livenessProbe:
-{{ toYaml .Values.clusterAgent.livenessProbe | indent 10 }}
+{{ $defaultLive := dict "httpGet" (dict "port" .Values.clusterAgent.healthPort "path" "/live" "scheme" "HTTP") }}
+{{ toYaml (mergeOverwrite $defaultLive .Values.clusterAgent.livenessProbe) | indent 10 }}
         readinessProbe:
-{{ toYaml .Values.clusterAgent.readinessProbe | indent 10 }}
+{{ $defaultReady := dict "httpGet" (dict "port" .Values.clusterAgent.healthPort "path" "/ready" "scheme" "HTTP") }}
+{{ toYaml (mergeOverwrite $defaultReady .Values.clusterAgent.readinessProbe) | indent 10 }}
         volumeMounts:
           - name: installinfo
             subPath: install_info

diff --git a/charts/datadog/templates/container-agent.yaml b/charts/datadog/templates/container-agent.yaml
@@ -165,7 +165,9 @@
 {{ toYaml .Values.agents.volumeMounts | indent 4 }}
 {{- end }}
   livenessProbe:
-{{ toYaml .Values.agents.containers.agent.livenessProbe | indent 4 }}
+{{ $defaultLive := dict "httpGet" (dict "port" .Values.agents.containers.agent.healthPort "path" "/live" "scheme" "HTTP") }}
+{{ toYaml (mergeOverwrite $defaultLive .Values.agents.containers.agent.livenessProbe) | indent 4 }}
   readinessProbe:
-{{ toYaml .Values.agents.containers.agent.readinessProbe | indent 4 }}
+{{ $defaultReady := dict "httpGet" (dict "port" .Values.agents.containers.agent.healthPort "path" "/ready" "scheme" "HTTP") }}
+{{ toYaml (mergeOverwrite $defaultReady .Values.agents.containers.agent.readinessProbe) | indent 4 }}
 {{- end -}}
diff --git a/charts/datadog/templates/container-trace-agent.yaml b/charts/datadog/templates/container-trace-agent.yaml
@@ -59,5 +59,6 @@
       mountPath: {{ (dir .Values.datadog.apm.socketPath) }}
     {{- end }}
   livenessProbe:
-{{ toYaml .Values.agents.containers.traceAgent.livenessProbe | indent 4 }}
+{{ $defaultLive := dict "tcpSocket" (dict "port" .Values.datadog.apm.port) }}
+{{ toYaml (mergeOverwrite $defaultLive .Values.agents.containers.traceAgent.livenessProbe) | indent 4 }}
 {{- end -}}
diff --git a/charts/datadog/values.yaml b/charts/datadog/values.yaml
@@ -427,29 +427,21 @@ clusterAgent:
   ## Ref: https://kubernetes.io/docs/concepts/configuration/assign-pod-node/#affinity-and-anti-affinity
   affinity: {}
 
-  # clusterAgent.healthPort -- Port number use the cluster-agent to server healthz endpoint
+  # clusterAgent.healthPort -- Port number to use in the Cluster Agent for the healthz endpoint
   healthPort: 5555
 
-  # clusterAgent.livenessProbe -- Override default agent liveness probe settings
+  # clusterAgent.livenessProbe -- Override default Cluster Agent liveness probe settings
   # @default -- Every 15s / 6 KO / 1 OK
   livenessProbe:
-    httpGet:
-      port: 5555
-      path: /live
-      scheme: HTTP
     initialDelaySeconds: 15
     periodSeconds: 15
     timeoutSeconds: 5
     successThreshold: 1
     failureThreshold: 6
 
-  # clusterAgent.readinessProbe -- Override default cluster-agent readiness probe settings
+  # clusterAgent.readinessProbe -- Override default Cluster Agent readiness probe settings
   # @default -- Every 15s / 6 KO / 1 OK
   readinessProbe:
-    httpGet:
-      port: 5555
-      path: /ready
-      scheme: HTTP
     initialDelaySeconds: 15
     periodSeconds: 15
     timeoutSeconds: 5
@@ -626,12 +618,12 @@ agents:
       #    cpu: 200m
       #    memory: 256Mi
 
+      # agents.containers.agent.healthPort -- Port number to use in the node agent for the healthz endpoint
+      healthPort: 5555
+
       # agents.containers.agent.livenessProbe -- Override default agent liveness probe settings
       # @default -- Every 15s / 6 KO / 1 OK
       livenessProbe:
-        httpGet:
-          path: /live
-          port: 5555
         initialDelaySeconds: 15
         periodSeconds: 15
         timeoutSeconds: 5
@@ -641,9 +633,6 @@ agents:
       # agents.containers.agent.readinessProbe -- Override default agent readiness probe settings
       # @default -- Every 15s / 6 KO / 1 OK
       readinessProbe:
-        httpGet:
-          path: /ready
-          port: 5555
         initialDelaySeconds: 15
         periodSeconds: 15
         timeoutSeconds: 5
@@ -685,8 +674,6 @@ agents:
       # agents.containers.traceAgent.livenessProbe -- Override default agent liveness probe settings
       # @default -- Every 15s
       livenessProbe:
-        tcpSocket:
-          port: 8126
         initialDelaySeconds: 15
         periodSeconds: 15
         timeoutSeconds: 5