Kubernetes

This guide will help you get LangFlow up and running in Kubernetes cluster, including the following steps:

• Install LangFlow as IDE in a Kubernetes cluster (for development)
• Install LangFlow as a standalone application in a Kubernetes cluster (for production runtime workloads)

LangFlow (IDE)

---

This solution is designed to provide a complete environment for developers to create, test, and debug their flows. It includes both the API and the UI.

Prerequisites

• Kubernetes server
• kubectl
• Helm

Step 0. Prepare a Kubernetes cluster

We use Minikube for this example, but you can use any Kubernetes cluster.

1. Create a Kubernetes cluster on Minikube.

   
   

   _10

   minikube start

   
   

2. Set kubectl to use Minikube.

   
   

   _10

   kubectl config use-context minikube

   
   

Step 1. Install the LangFlow Helm chart

1. Add the repository to Helm.

   
   

   _10

   helm repo add langflow <https://langflow-ai.github.io/langflow-helm-charts>

   _10

   helm repo update

   
   

2. Install LangFlow with the default options in the langflow namespace.

   
   

   _10

   helm install langflow-ide langflow/langflow-ide -n langflow --create-namespace

   
   

3. Check the status of the pods

   
   

   _10

   kubectl get pods -n langflow

   
   
   
   

   _10

   NAME READY STATUS RESTARTS AGE

   _10

   langflow-0 1/1 Running 0 33s

   _10

   langflow-frontend-5d9c558dbb-g7tc9 1/1 Running 0 38s

   
   

Step 2. Access LangFlow

Enable local port forwarding to access LangFlow from your local machine.

_10

kubectl port-forward -n langflow svc/langflow-langflow-runtime 7860:7860

Now you can access LangFlow at http://localhost:7860/.

LangFlow version

To specify a different LangFlow version, you can set the langflow.backend.image.tag and langflow.frontend.image.tag values in the values.yaml file.

_10

langflow:

_10

backend:

_10

image:

_10

tag: "1.0.0a59"

_10

frontend:

_10

image:

_10

tag: "1.0.0a59"

Storage

By default, the chart will use a SQLLite database stored in a local persistent disk. If you want to use an external PostgreSQL database, you can set the langflow.database values in the values.yaml file.

_30

# Deploy postgresql. You can skip this section if you have an existing postgresql database.

_30

postgresql:

_30

enabled: true

_30

fullnameOverride: "langflow-ide-postgresql-service"

_30

auth:

_30

username: "langflow"

_30

password: "langflow-postgres"

_30

database: "langflow-db"

_30

_30

langflow:

_30

backend:

_30

externalDatabase:

_30

enabled: true

_30

driver:

_30

value: "postgresql"

_30

host:

_30

value: "langflow-ide-postgresql-service"

_30

port:

_30

value: "5432"

_30

database:

_30

value: "langflow-db"

_30

user:

_30

value: "langflow"

_30

password:

_30

valueFrom:

_30

secretKeyRef:

_30

key: "password"

_30

name: "langflow-ide-postgresql-service"

_30

sqlite:

_30

enabled: false

Scaling

You can scale the number of replicas for the LangFlow backend and frontend services by changing the replicaCount value in the values.yaml file.

_10

langflow:

_10

backend:

_10

replicaCount: 3

_10

frontend:

_10

replicaCount: 3

You can scale frontend and backend services independently.

To scale vertically (increase the resources for the pods), you can set the resources values in the values.yaml file.

_11

langflow:

_11

backend:

_11

resources:

_11

requests:

_11

memory: "2Gi"

_11

cpu: "1000m"

_11

frontend:

_11

resources:

_11

requests:

_11

memory: "1Gi"

_11

cpu: "1000m"

Deploy on AWS EKS, Google GKE, or Azure AKS and other examples

Visit the LangFlow Helm Charts repository for more information.

LangFlow (Runtime)

---

The runtime chart is tailored for deploying applications in a production environment. It is focused on stability, performance, isolation, and security to ensure that applications run reliably and efficiently.

Using a dedicated deployment for a set of flows is fundamental in production environments to have granular resource control.

Prerequisites

• Kubernetes server
• kubectl
• Helm

Step 0. Prepare a Kubernetes cluster

Follow the same steps as for the LangFlow IDE.

Step 1. Install the LangFlow runtime Helm chart

1. Add the repository to Helm.

   
   

   _10

   helm repo add langflow <https://langflow-ai.github.io/langflow-helm-charts>

   _10

   helm repo update

   
   

2. Install the LangFlow app with the default options in the langflow namespace. If you bundled the flow in a docker image, you can specify the image name in the values.yaml file or with the -set flag: If you want to download the flow from a remote location, you can specify the URL in the values.yaml file or with the -set flag:

   
   

   _10

   helm install my-langflow-app langflow/langflow-runtime -n langflow --create-namespace --set image.repository=myuser/langflow-just-chat --set image.tag=1.0.0

   
   
   
   

   _10

   helm install my-langflow-app langflow/langflow-runtime -n langflow --create-namespace --set downloadFlows.flows[0].url=https://raw.githubusercontent.com/langflow-ai/langflow/dev/src/backend/base/langflow/initial_setup/starter_projects/Basic%20Prompting%20(Hello%2C%20world!).json

   
   

3. Check the status of the pods.

   
   

   _10

   kubectl get pods -n langflow

   
   

Step 2. Access the LangFlow app API

Enable local port forwarding to access LangFlow from your local machine.

_10

kubectl port-forward -n langflow svc/langflow-my-langflow-app 7860:7860

Now you can access the API at http://localhost:7860/api/v1/flows and execute the flow:

_10

id=$(curl -s <http://localhost:7860/api/v1/flows> | jq -r '.flows[0].id')

_10

curl -X POST \\

_10

"<http://localhost:7860/api/v1/run/$id?stream=false>" \\

_10

-H 'Content-Type: application/json'\\

_10

-d '{

_10

"input_value": "Hello!",

_10

"output_type": "chat",

_10

"input_type": "chat"

_10

}'

Storage

In this case, storage is not needed as our deployment is stateless.

Log level and LangFlow configurations

You can set the log level and other LangFlow configurations in the values.yaml file.

_10

env:

_10

- name: LANGFLOW_LOG_LEVEL

_10

value: "INFO"

Configure secrets and variables

To inject secrets and LangFlow global variables, you can use the secrets and env sections in the values.yaml file.

Let's say your flow uses a global variable which is a secret; when you export the flow as JSON, it's recommended to not include it. When importing the flow in the LangFlow runtime, you can set the global variable using the env section in the values.yaml file. Assuming you have a global variable called openai_key_var, you can read it directly from a secret:

_10

env:

_10

- name: openai_key_var

_10

valueFrom:

_10

secretKeyRef:

_10

name: openai-key

_10

key: openai-key

or directly from the values file (not recommended for secret values!):

_10

env:

_10

- name: openai_key_var

_10

value: "sk-...."

Scaling

You can scale the number of replicas for the LangFlow app by changing the replicaCount value in the values.yaml file.

_10

replicaCount: 3

To scale vertically (increase the resources for the pods), you can set the resources values in the values.yaml file.

_10

resources:

_10

requests:

_10

memory: "2Gi"

_10

cpu: "1000m"

Other Examples

---

Visit the LangFlow Helm Charts repository for more examples and configurations. Use the default values file as reference for all the options available.

note

Visit the examples directory to learn more about different deployment options.