feat(chart): add optional startupProbe for web and worker

charts/langfuse/Chart.yaml

@@ -1,6 +1,6 @@
 apiVersion: v2
 name: langfuse
-version: 1.5.35
+version: 1.6.0
 description: Open source LLM engineering platform - LLM observability, metrics, evaluations, prompt management.
 type: application
 keywords:

charts/langfuse/README.md

@@ -1,6 +1,6 @@
 # langfuse
 
-![Version: 1.5.35](https://img.shields.io/badge/Version-1.5.35-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 3.185.0](https://img.shields.io/badge/AppVersion-3.185.0-informational?style=flat-square)
+![Version: 1.6.0](https://img.shields.io/badge/Version-1.6.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 3.185.0](https://img.shields.io/badge/AppVersion-3.185.0-informational?style=flat-square)
 
 Open source LLM engineering platform - LLM observability, metrics, evaluations, prompt management.
 
@@ -155,6 +155,13 @@ Open source LLM engineering platform - LLM observability, metrics, evaluations,
 | langfuse.web.service.nodePort | string | `nil` | The node port to use for the langfuse web application |
 | langfuse.web.service.port | int | `3000` | The port to use for the langfuse web application |
 | langfuse.web.service.type | string | `"ClusterIP"` | The type of service to use for the langfuse web application |
+| langfuse.web.startupProbe.enabled | bool | `false` | Set to `true` to enable the startupProbe for the langfuse web pods. |
+| langfuse.web.startupProbe.failureThreshold | int | `30` | Failure threshold for startupProbe. With the default periodSeconds this allows ~5 minutes for startup. |
+| langfuse.web.startupProbe.initialDelaySeconds | int | `0` | Initial delay seconds for startupProbe. |
+| langfuse.web.startupProbe.path | string | `"/api/public/health"` | Path to check for startup. |
+| langfuse.web.startupProbe.periodSeconds | int | `10` | Period seconds for startupProbe. |
+| langfuse.web.startupProbe.successThreshold | int | `1` | Success threshold for startupProbe. |
+| langfuse.web.startupProbe.timeoutSeconds | int | `5` | Timeout seconds for startupProbe. |
 | langfuse.web.vpa.controlledResources | list | `[]` | The resources to control for the langfuse web pods |
 | langfuse.web.vpa.enabled | bool | `false` | Set to `true` to enable VPA for the langfuse web pods |
 | langfuse.web.vpa.maxAllowed | object | `{}` | The maximum allowed resources for the langfuse web pods |
@@ -201,6 +208,12 @@ Open source LLM engineering platform - LLM observability, metrics, evaluations,
 | langfuse.worker.replicas | string | `nil` | Number of replicas to use if HPA is not enabled. Defaults to the global replicas |
 | langfuse.worker.resources | object | `{}` | Resources for the langfuse worker pods. Defaults to the global resources |
 | langfuse.worker.revisionHistoryLimit | string | `nil` | Number of old ReplicaSets to retain to allow rollback. |
+| langfuse.worker.startupProbe.enabled | bool | `false` | Set to `true` to enable the startupProbe for the langfuse worker pods. |
+| langfuse.worker.startupProbe.failureThreshold | int | `30` | Failure threshold for startupProbe. With the default periodSeconds this allows ~5 minutes for startup. |
+| langfuse.worker.startupProbe.initialDelaySeconds | int | `0` | Initial delay seconds for startupProbe. |
+| langfuse.worker.startupProbe.periodSeconds | int | `10` | Period seconds for startupProbe. |
+| langfuse.worker.startupProbe.successThreshold | int | `1` | Success threshold for startupProbe. |
+| langfuse.worker.startupProbe.timeoutSeconds | int | `5` | Timeout seconds for startupProbe. |
 | langfuse.worker.vpa.controlledResources | list | `[]` | The resources to control for the langfuse worker pods |
 | langfuse.worker.vpa.enabled | bool | `false` | Set to `true` to enable VPA for the langfuse worker pods |
 | langfuse.worker.vpa.maxAllowed | object | `{}` | The maximum allowed resources for the langfuse worker pods |

charts/langfuse/templates/web/deployment.yaml

@@ -93,6 +93,17 @@ spec:
             - name: http
               containerPort: {{ .Values.langfuse.web.service.port }}
               protocol: TCP
+          {{- if .Values.langfuse.web.startupProbe.enabled }}
+          startupProbe:
+            httpGet:
+              path: {{ .Values.langfuse.web.startupProbe.path | default "/api/public/health" }}
+              port: http
+            initialDelaySeconds: {{ .Values.langfuse.web.startupProbe.initialDelaySeconds | default 0 }}
+            periodSeconds: {{ .Values.langfuse.web.startupProbe.periodSeconds | default 10 }}
+            timeoutSeconds: {{ .Values.langfuse.web.startupProbe.timeoutSeconds | default 5 }}
+            successThreshold: {{ .Values.langfuse.web.startupProbe.successThreshold | default 1 }}
+            failureThreshold: {{ .Values.langfuse.web.startupProbe.failureThreshold | default 30 }}
+          {{- end }}
           livenessProbe:
             httpGet:
               path: {{ .Values.langfuse.web.livenessProbe.path | default "/api/public/health" }}

charts/langfuse/templates/worker/deployment.yaml

@@ -91,6 +91,17 @@ spec:
             - name: http
               containerPort: {{ .Values.langfuse.worker.port | default 3030 }}
               protocol: TCP
+          {{- if .Values.langfuse.worker.startupProbe.enabled }}
+          startupProbe:
+            httpGet:
+              path: /api/health
+              port: http
+            initialDelaySeconds: {{ .Values.langfuse.worker.startupProbe.initialDelaySeconds | default 0 }}
+            periodSeconds: {{ .Values.langfuse.worker.startupProbe.periodSeconds | default 10 }}
+            timeoutSeconds: {{ .Values.langfuse.worker.startupProbe.timeoutSeconds | default 5 }}
+            successThreshold: {{ .Values.langfuse.worker.startupProbe.successThreshold | default 1 }}
+            failureThreshold: {{ .Values.langfuse.worker.startupProbe.failureThreshold | default 30 }}
+          {{- end }}
           livenessProbe:
             httpGet:
               path: /api/health

charts/langfuse/tests/startup-probe_test.yaml

@@ -0,0 +1,47 @@
+suite: test startupProbe configuration
+templates:
+  - web/deployment.yaml
+  - worker/deployment.yaml
+tests:
+  - it: should not render a startupProbe by default
+    values:
+      - ../values.lint.yaml
+    asserts:
+      - notExists:
+          path: spec.template.spec.containers[0].startupProbe
+        template: web/deployment.yaml
+      - notExists:
+          path: spec.template.spec.containers[0].startupProbe
+        template: worker/deployment.yaml
+
+  - it: should render the web startupProbe when enabled
+    values:
+      - ../values.lint.yaml
+    set:
+      langfuse.web.startupProbe.enabled: true
+    asserts:
+      - exists:
+          path: spec.template.spec.containers[0].startupProbe
+        template: web/deployment.yaml
+      - equal:
+          path: spec.template.spec.containers[0].startupProbe.httpGet.path
+          value: /api/public/health
+        template: web/deployment.yaml
+      - equal:
+          path: spec.template.spec.containers[0].startupProbe.failureThreshold
+          value: 30
+        template: web/deployment.yaml
+
+  - it: should render the worker startupProbe when enabled
+    values:
+      - ../values.lint.yaml
+    set:
+      langfuse.worker.startupProbe.enabled: true
+    asserts:
+      - exists:
+          path: spec.template.spec.containers[0].startupProbe
+        template: worker/deployment.yaml
+      - equal:
+          path: spec.template.spec.containers[0].startupProbe.httpGet.path
+          value: /api/health
+        template: worker/deployment.yaml

charts/langfuse/values.yaml

@@ -307,6 +307,23 @@ langfuse:
       # -- Success threshold for readinessProbe.
       successThreshold: 1
 
+    # Startup probe configuration for the web deployment
+    startupProbe:
+      # -- Set to `true` to enable the startupProbe for the langfuse web pods.
+      enabled: false
+      # -- Path to check for startup.
+      path: "/api/public/health"
+      # -- Initial delay seconds for startupProbe.
+      initialDelaySeconds: 0
+      # -- Period seconds for startupProbe.
+      periodSeconds: 10
+      # -- Timeout seconds for startupProbe.
+      timeoutSeconds: 5
+      # -- Failure threshold for startupProbe. With the default periodSeconds this allows ~5 minutes for startup.
+      failureThreshold: 30
+      # -- Success threshold for startupProbe.
+      successThreshold: 1
+
     # Pod Disruption Budget configuration for the web deployment
     pdb:
       # -- Set to `true` to create a Pod Disruption Budget for the langfuse web pods
@@ -433,6 +450,21 @@ langfuse:
       # -- Success threshold for livenessProbe.
       successThreshold: 1
 
+    # Startup probe configuration for the worker deployment
+    startupProbe:
+      # -- Set to `true` to enable the startupProbe for the langfuse worker pods.
+      enabled: false
+      # -- Initial delay seconds for startupProbe.
+      initialDelaySeconds: 0
+      # -- Period seconds for startupProbe.
+      periodSeconds: 10
+      # -- Timeout seconds for startupProbe.
+      timeoutSeconds: 5
+      # -- Failure threshold for startupProbe. With the default periodSeconds this allows ~5 minutes for startup.
+      failureThreshold: 30
+      # -- Success threshold for startupProbe.
+      successThreshold: 1
+
   # NextAuth configuration
   nextauth:
     # -- When deploying to production, set the `nextauth.url` value to the canonical URL of your site.