This Helm chart deploys the BotForge application, including the Backend, RAG service, Redis, and PostgreSQL.
- Kubernetes cluster (e.g., k3s, minikube, EKS, GKE)
- Helm installed
- Docker images for
botforge-backendandbotforge-ragavailable in your cluster or registry.
-
Build Images (if running locally/custom):
# Example for local k3s/minikube docker build -t botforge-backend:latest ../botforge-backend docker build -t botforge-rag:latest ../botforge-rag # Import to cluster if needed (e.g., k3s) k3s ctr images import botforge-backend.tar
-
Configure Secrets: Edit
values.yamlto set your actual secrets, or pass them via--set.rag.env.OPENAI_API_KEYrag.env.UPSTASH_TOKENbackend.config.jwtSecretbackend.config.firebase.apiKey
-
Install Chart:
helm install botforge ./botforge-helm --namespace botforge --create-namespace
The values.yaml file contains the default configuration.
- Backend: NodePort 30080 (default). Access at
http://<node-ip>:30080. - RAG: ClusterIP. Internal communication only.
- Redis: ClusterIP. Internal.
- Postgres: ClusterIP. Internal.
- Postgres: Uses
emptyDirby default. For production, configure a PVC indeployment-postgres.yaml. - Redis: Uses
emptyDirby default. - Uploads: Uses
emptyDirby default.
The chart is pre-configured with the environment variables provided for the RAG service, including:
OPENAI_API_KEYUPSTASH_URL/UPSTASH_TOKENPOSTGRES_*(configured to use the internal Postgres)REDIS_*(configured to use the internal Redis)