Tock Helm Chart initial version for theopenconversationkit (#1)

* initial version

* updates

readme,
upgrade tock version  to 24.9.3
solve tock configuration issues

* upgrade genai orchestrator configs

* add pgvector , upgrade mongodb helmchart

* update helper for pgvector

* add rag sample

.gitignore

@@ -0,0 +1,8 @@
+charts/tock/README-Origin.md
+.DS_Store
+/tmp
+
+### Helm ###
+# Chart dependencies
+**/charts/*.tgz
+packages/

Makefile

@@ -0,0 +1,29 @@
+CHART=charts/tock
+chartversion?=`awk '/^version/ {print $$NF}' ${CHART}/Chart.yaml`
+appversion?=`awk '/^appVersion/ {print $$NF}' ${CHART}/Chart.yaml`
+
+
+lint:
+	helm lint ${CHART}
+
+debug:
+	helm install --dry-run --debug tockdebug ${CHART}
+
+template:
+	helm template test ${CHART} --debug
+
+version:
+	@echo "Chart Name:tock Application Version:$(appversion)  Chart Version:${chartversion}"
+
+build:
+	helm repo add bitnami https://charts.bitnami.com/bitnami
+	helm dep update ${CHART}
+	helm lint ${CHART}
+	echo "Chart Name:Tock Application Version:${appversion}  Chart Version:${chartversion} "
+	helm package ${CHART} --version ${chartversion} --app-version ${appversion} --destination packages
+
+doc:
+	helm-docs -c ${CHART}
+
+publish:
+	helm push packages/tock-${chartversion}.tgz oci://registry.hub.docker.com/onelans

README.md

@@ -0,0 +1,27 @@
+# Helm Chart for Tock  (theopenconversationkit)
+
+
+[Tock](https://doc.tock.ai/fr/), an open conversational AI platform provide a complete solution to build conversational agents aka bots.
+
+Tock can integrate and experiment with both classic and Generative AI (LLM, RAG) models.
+
+
+## Quick start
+
+```console 
+$ helm install mytock ./charts/tock
+```
+
+## DLDR
+
+```console
+$ helm install my-release oci://registry.hub.docker.com/onelans/tock --version 0.4.4
+```
+
+You will find more information on chart parameters at the helm chart [README](charts/tock/README.md).
+
+## What is deployed 
+
+The chart deploy all tock components. It's able also to deploy a mongodb database (Bitnami mongoDb chart is used as subchart) or use an existing mongodb backend. In the same feeling, since version 24.3.2 of tock, opensearch can be deployed as sub chart (opensearch chart is used as subchart) and since version 24.9.3 PGvector can be deployer as sub chart (bitnami chart is used as subchart).
+
+![Tock on K8S](tock-24x-on-k8s.png)

charts/tock/Chart.lock

@@ -0,0 +1,15 @@
+dependencies:
+- name: mongodb
+  repository: https://charts.bitnami.com/bitnami
+  version: 14.8.3
+- name: opensearch
+  repository: https://opensearch-project.github.io/helm-charts/
+  version: 2.21.0
+- name: postgresql
+  repository: https://charts.bitnami.com/bitnami
+  version: 13.4.4
+- name: opensearch-dashboards
+  repository: https://opensearch-project.github.io/helm-charts/
+  version: 2.19.0
+digest: sha256:2e07cc09c0ec22303869433fbefe49c66b34d643bf49d41e018a38b65befb8ac
+generated: "2024-10-23T08:49:04.181528+02:00"

charts/tock/Chart.yaml

@@ -0,0 +1,40 @@
+apiVersion: v2
+appVersion: 24.9.3
+description: A helm chart for Tock. Tock is an open conversational AI platform. It's a complete solution to build conversational agents aka bots.Tock can integrate and experiment with both classic and Generative AI (LLM, RAG) models
+name: tock
+type: application
+version: 0.4.4
+
+maintainers:
+  - name: stephanacquatella
+
+dependencies:
+- name: mongodb
+  #version: 13.6.8
+  version: 14.8.3
+  repository : https://charts.bitnami.com/bitnami
+  condition: global.deployMongoDb.enabled
+  tags: 
+    - mongodb
+
+- name: opensearch
+  version: 2.21.0
+  repository : https://opensearch-project.github.io/helm-charts/
+  condition: global.deployOpenSearch.enabled
+  tags: 
+    - opensearch
+
+- name: postgresql
+  version: 13.4.4
+  repository : https://charts.bitnami.com/bitnami
+  condition: global.deployPgVector.enabled
+  tags: 
+    - postgres 
+
+- name: opensearch-dashboards
+  version: 2.19.0
+  repository : https://opensearch-project.github.io/helm-charts/
+  condition: global.deployOpenSearch.enabled
+  tags: 
+    - opensearch
+    - dashboard

charts/tock/README.md.gotmpl

@@ -0,0 +1,160 @@
+{{ template "chart.header" . }}
+{{ template "chart.description" . }}
+
+{{ template "chart.versionBadge" . }}{{ template "chart.typeBadge" . }}{{ template "chart.appVersionBadge" . }}
+
+
+
+## DLDR
+
+To install the chart with the release name `my-release`:
+
+```console
+$ helm registry login -u myuser registry.hub.docker.com 
+$ helm install my-release  oci://registry.hub.docker.com/onelans/{{ template "chart.name" . }} --version 0.4.4
+```
+
+## Introduction
+
+This chart helps to setup a Tock environnement. 
+
+
+{{ template "chart.requirementsSection" . }}
+
+## Sections
+
+This creates values, but sectioned into own section tables if a section comment is provided.
+
+{{ template "chart.valuesSection" . }}
+
+
+
+## Authentification configurations
+
+The following sample could be added as ConfigMap to configure the authentication of the admin web interface.
+
+To use it , you have to apply the following ConfigMap to your cluster.
+
+```console
+$ kubectl apply -f ./admin-web-auth-cfg.yaml
+```
+
+And set in your values file `adminWeb.authCongifMap` to `admin-web-auth-cfg` (configmap name)
+
+`admin-web-auth-cfg.yaml`
+
+```yaml
+apiVersion: v1
+  kind: ConfigMap
+  metadata:
+    name: admin-web-auth-cfg
+    labels:
+      app.kubernetes.io/name: admin-web-auth
+      app.kubernetes.io/component: admin-web
+    data:
+      tock_users:  "alice@tock.ai,bob@tock.ai" # Credentials (separated by commas). Default value `admin@app.com`
+      tock_passwords: "secret1,secret2" # Password (separated by commas). Default value `password`
+      tock_organizations: "tock,tock" # Organizations (separated by commas). Default value `app``
+      tock_roles: "botUser,nlpUser|botUser|admin|technicalAdmin" #  Roles separated | (and then by commas). Default value is empty."
+ ```
+
+In this example, Alice has the role 'botUser', whereas Bob has all roles. 
+To define the identities and roles of several users, separate their values with commas.
+
+You can find more information about the roles in the [Tock documentation](https://doc.tock.ai/tock/fr/admin/securite/#r%C3%B4les)
+
+## Deployment on arm64 and processor without AVX instructions
+
+It seems the native build of MongoDB requires AVX instructions at the processor level
+
+https://github.com/bitnami/charts/issues/12834
+
+For Arm, the image used in value must be changed and the following Mongodb chart image should be used instead.
+
+https://artifacthub.io/packages/helm/bitnami/mongodb/14.8.3
+
+## Deployment sample on Rancher Desktop or k3s
+
+```console
+$ helm install mytock oci://registry.hub.docker.com/onelans/{{ template "chart.name" . }} --version 0.4.4 -f ./rancher-values.yaml
+```
+
+`rancher-values.yaml`
+```yaml
+global:
+  wildcardDomain: rancher.localhost
+  deployMongoDb:
+    enabled: true
+
+botApi:
+  environment:
+    tock_default_log_level: "debug"
+    tock_web_use_default_cors_handler_url : "*"
+  ingress:
+    enabled: true
+
+
+mongodb:
+  architecture: "replicaset"
+  auth:
+    enabled: false
+  persistence:
+    enabled: false
+    size: 1Gi
+  image:
+    repository: xavidop/mongodb
+    tag: 7.0.14
+
+nlpApi:
+  environment:
+    tock_env: prod
+    tock_web_use_default_cors_handler: "true"
+    tock_web_use_default_cors_handler_url: "*"
+    tock_web_use_default_cors_handler_with_credentials: "false"
+  repository:
+    image: tock/nlp_api
+    tag: 24.3.4
+
+
+adminWeb:
+  ingress:
+    enabled: true
+```
+
+> The values file defines a MongoDB image that is compatible with macOS Arm
+
+## Deployment sample on GKE
+
+```console
+$ helm install mytock oci://registry.hub.docker.com/onelans/{{ template "chart.name" . }} --version 0.4.4 -f ./gke-values.yaml
+```
+
+`gke-values.yaml`
+
+```yaml
+global:
+  wildcardDomain: gke.mydomain.com
+  deployMongoDb:
+    enabled: true
+
+botApi:
+  ingress:
+    enabled: true
+    annotations:
+      kubernetes.io/ingress.class: "gce"
+  service:
+    type: NodePort
+
+adminWeb:
+  ingress:
+    enabled: true
+    annotations:
+      kubernetes.io/ingress.class: "gce"  
+  service:
+    type: NodePort
+```
+
+> The values file defines the use of the GCE ingress controller
+> You can get the external IP of the ingress controller with the following command
+> `kubectl get ingress mytock-admin-web  --output yaml`
+

charts/tock/templates/NOTES.txt

@@ -0,0 +1,46 @@
+
+Congratulations,
+
+Your application tock is now deployed. This may take a while to be up and responding.
+
+Get the Tock Studio URL by running these commands:
+
+{{- if .Values.adminWeb.ingress.enabled }}
+http{{ if .Values.adminWeb.ingress.tls }}s{{ end }}://tockstudio-{{ .Release.Name }}.{{ .Values.global.wildcardDomain }}{{ .Values.adminWeb.ingress.path }}
+{{- else if contains "NodePort" .Values.adminWeb.service.type }}
+  export NODE_PORT=$(kubectl get --namespace {{ .Release.Namespace }} -o jsonpath="{.spec.ports[0].nodePort}" services {{ .Release.Name }}-admin-web)
+  export NODE_IP=$(kubectl get nodes --namespace {{ .Release.Namespace }} -o jsonpath="{.items[0].status.addresses[0].address}")
+  echo http://$NODE_IP:$NODE_PORT
+{{- else if contains "LoadBalancer" .Values.adminWeb.service.type }}
+     NOTE: It may take a few minutes for the LoadBalancer IP to be available.
+           You can watch the status of by running 'kubectl get svc -w {{ .Release.Name }}-admin-web'
+  export SERVICE_IP=$(kubectl get svc --namespace {{ .Release.Namespace }} {{ .Release.Name }}-admin-web -o jsonpath='{.status.loadBalancer.ingress[0].ip}')
+  echo http://$SERVICE_IP:{{ .Values.adminWeb.service.port }}
+{{- else if contains "ClusterIP" .Values.adminWeb.service.type }}
+  export POD_NAME=$(kubectl get pods --namespace {{ .Release.Namespace }} -l "app.kubernetes.io/component=admin-web,app.kubernetes.io/instance={{ .Release.Name }}" -o jsonpath="{.items[0].metadata.name}")
+  echo "Visit http://127.0.0.1:8080 to use your application"
+  kubectl port-forward $POD_NAME 8080:8080
+{{- end }}
+
+
+The default login/password is admin@app.com/password. 
+
+
+Bot api can be reach through the following url:
+
+{{- if .Values.botApi.ingress.enabled }}
+http{{ if .Values.botApi.ingress.tls }}s{{ end }}://bot-api-{{ .Release.Name }}.{{ .Values.global.wildcardDomain }}{{ .Values.botApi.ingress.path }}
+{{- else if contains "NodePort" .Values.botApi.service.type }}
+  export NODE_PORT=$(kubectl get --namespace {{ .Release.Namespace }} -o jsonpath="{.spec.ports[0].nodePort}" services  {{ .Release.Name }}-bot-api)
+  export NODE_IP=$(kubectl get nodes --namespace {{ .Release.Namespace }} -o jsonpath="{.items[0].status.addresses[0].address}")
+  echo http://$NODE_IP:$NODE_PORT
+{{- else if contains "LoadBalancer" .Values.botApi.service.type }}
+     NOTE: It may take a few minutes for the LoadBalancer IP to be available.
+           You can watch the status of by running 'kubectl get svc -w {{ .Release.Name }}-bot-api'
+  export SERVICE_IP=$(kubectl get svc --namespace {{ .Release.Namespace }} {{ .Release.Name }}-bot-api -o jsonpath='{.status.loadBalancer.ingress[0].ip}')
+  echo http://$SERVICE_IP:{{ .Values.adminWeb.service.port }}
+{{- else if contains "ClusterIP" .Values.botApi.service.type }}
+  export POD_NAME=$(kubectl get pods --namespace {{ .Release.Namespace }} -l "app.kubernetes.io/component=bot-api,app.kubernetes.io/instance={{ .Release.Name }}" -o jsonpath="{.items[0].metadata.name}")
+  echo "Visit http://127.0.0.1:8080 to use your application"
+  kubectl port-forward $POD_NAME 8080:8080
+{{- end }}