Giters

langflow-ai / langflow

⛓️ Langflow is a visual framework for building multi-agent and RAG applications. It's open-source, Python-powered, fully customizable, model and vector store agnostic.

Home Page:http://www.langflow.org

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

No Response when running the Flow using cURL or Python API

edwinjosechittilappilly opened this issue · comments

commented

Describe the bug
No Response when running the Flow using cURL or Python API.

Even tried to ping through the terminal. The server is not reachable/ gives back unknown host.

Requesting more Documentation on Deployment of langflow apps.

commented

To resolve the issue where there is no response when running the Flow using cURL or the Python API, and attempts to ping the server result in an "unknown host" error, you can follow the documentation provided for deploying Langflow apps in backend-only mode or on Google Cloud Platform (GCP).

Here are the steps to ensure proper deployment and interaction with the Langflow server:

  1. Start Langflow in Backend-Only Mode:

    python3 -m langflow run --backend-only

    This command will start Langflow in backend-only mode, exposing the API without running the frontend UI. The terminal should print Welcome to ⛓ Langflow, and a blank window should open at http://127.0.0.1:7864/all.

  2. Download Your Flow's cURL Call:

    • Click on API in the Langflow UI.
    • Click curl > Copy code and save the code to your local machine. The cURL command will look something like this: 
      curl -X POST \
    "http://127.0.0.1:7864/api/v1/run/ef7e0554-69e5-4e3e-ab29-ee83bcd8d9ef?stream=false" \
    -H 'Content-Type: application/json' \
    -d '{"input_value": "message", "output_type": "chat", "input_type": "chat", "tweaks": {"Prompt-kvo86": {}, "OpenAIModel-MilkD": {}, "ChatOutput-ktwdw": {}, "ChatInput-xXC4F": {}}}'

  3. Run the cURL Command:

    curl -X POST \
    "http://127.0.0.1:7864/api/v1/run/ef7e0554-69e5-4e3e-ab29-ee83bcd8d9ef?stream=false" \
    -H 'Content-Type: application/json' \
    -d '{"input_value": "message", "output_type": "chat", "input_type": "chat", "tweaks": {"Prompt-kvo86": {}, "OpenAIModel-MilkD": {}, "ChatOutput-ktwdw": {}, "ChatInput-xXC4F": {}}}'

  4. Download Your Flow's Python API Call:

    • Click on API in the Langflow UI.
    • Click Python API > Copy code and save the code to your local machine. The Python code will look something like this: 
      import requests
from typing import Optional

BASE_API_URL = "http://127.0.0.1:7864/api/v1/run"
FLOW_ID = "ef7e0554-69e5-4e3e-ab29-ee83bcd8d9ef"

def run_flow(message: str, flow_id: str, output_type: str = "chat", input_type: str = "chat", tweaks: Optional[dict] = None, api_key: Optional[str] = None) -> dict:
    api_url = f"{BASE_API_URL}/{flow_id}"
    payload = {
        "input_value": message,
        "output_type": output_type,
        "input_type": input_type,
    }
    headers = None
    if tweaks:
        payload["tweaks"] = tweaks
    if api_key:
        headers = {"x-api-key": api_key}
    response = requests.post(api_url, json=payload, headers=headers)
    return response.json()

message = "message"
print(run_flow(message=message, flow_id=FLOW_ID))

  5. Run Your Python App:

    python3 app.py

  6. Check Your Environment Configuration:
    Ensure that your .env file is correctly configured. Here is an example configuration:

    DOMAIN=localhost
STACK_NAME=langflow-stack
ENVIRONMENT=development

TRAEFIK_PUBLIC_NETWORK=traefik-public
TRAEFIK_TAG=langflow-traefik
TRAEFIK_PUBLIC_TAG=traefik-public

# RabbitMQ configuration
RABBITMQ_DEFAULT_USER=langflow
RABBITMQ_DEFAULT_PASS=langflow

# Database configuration
DB_USER=langflow
DB_PASSWORD=langflow
DB_HOST=db
DB_PORT=5432
DB_NAME=langflow

# Logging configuration
LOG_LEVEL=debug

# DB configuration
POSTGRES_USER=langflow
POSTGRES_PASSWORD=langflow
POSTGRES_DB=langflow
POSTGRES_PORT=5432

# Flower configuration
LANGFLOW_CACHE_TYPE=redis
LANGFLOW_REDIS_HOST=result_backend
LANGFLOW_REDIS_PORT=6379
LANGFLOW_REDIS_DB=0
LANGFLOW_REDIS_EXPIRE=3600
LANGFLOW_REDIS_PASSWORD=
FLOWER_UNAUTHENTICATED_API=True
BROKER_URL=amqp://langflow:langflow@broker:5672
RESULT_BACKEND=redis://result_backend:6379/0
C_FORCE_ROOT="true"

# Frontend configuration
VITE_PROXY_TARGET=http://backend:7860/api/
BACKEND_URL=http://backend:7860

# PGAdmin configuration
PGADMIN_DEFAULT_EMAIL=admin@admin.com
PGADMIN_DEFAULT_PASSWORD=admin

# OpenAI configuration (for testing purposes)
OPENAI_API_KEY=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

# Superuser configuration
LANGFLOW_SUPERUSER=superuser
LANGFLOW_SUPERUSER_PASSWORD=superuser

# New user configuration
LANGFLOW_NEW_USER_IS_ACTIVE=False

  7. Deploy Langflow on Google Cloud Platform (GCP):
    If you are deploying Langflow on GCP, you can use the following script to set up the environment and deploy the application:

    # Set the VM, image, and networking configuration
VM_NAME="langflow-dev"
IMAGE_FAMILY="debian-11"
IMAGE_PROJECT="debian-cloud"
BOOT_DISK_SIZE="100GB"
ZONE="us-central1-a"
REGION="us-central1"
VPC_NAME="default"
SUBNET_NAME="default"
SUBNET_RANGE="10.128.0.0/20"
NAT_GATEWAY_NAME="nat-gateway"
CLOUD_ROUTER_NAME="nat-client"

# Set the GCP project's compute region
gcloud config set compute/region $REGION

# Check if the VPC exists, and create it if not
vpc_exists=$(gcloud compute networks list --filter="name=$VPC_NAME" --format="value(name)")
if [[ -z "$vpc_exists" ]]; then
  gcloud compute networks create $VPC_NAME --subnet-mode=custom
fi

# Check if the subnet exists, and create it if not
subnet_exists=$(gcloud compute networks subnets list --filter="name=$SUBNET_NAME AND region=$REGION" --format="value(name)")
if [[ -z "$subnet_exists" ]]; then
  gcloud compute networks subnets create $SUBNET_NAME --network=$VPC_NAME --region=$REGION --range=$SUBNET_RANGE
fi

# Create a firewall rule to allow TCP port 7860 for all instances in the VPC
firewall_7860_exists=$(gcloud compute firewall-rules list --filter="name=allow-tcp-7860" --format="value(name)")
if [[ -z "$firewall_7860_exists" ]]; then
  gcloud compute firewall-rules create allow-tcp-7860 --network $VPC_NAME --allow tcp:7860 --source-ranges 0.0.0.0/0 --direction INGRESS
fi

# Create a firewall rule to allow IAP traffic
firewall_iap_exists=$(gcloud compute firewall-rules list --filter="name=allow-iap" --format="value(name)")
if [[ -z "$firewall_iap_exists" ]]; then
    gcloud compute firewall-rules create allow-iap --network $VPC_NAME --allow tcp:80,tcp:443,tcp:22,tcp:3389 --source-ranges 35.235.240.0/20 --direction INGRESS
fi

# Define the startup script as a multiline Bash here-doc
STARTUP_SCRIPT=$(cat <<'EOF'
#!/bin/bash

# Update and upgrade the system
apt -y update
apt -y upgrade

# Install Python 3 pip, Langflow, and Nginx
apt -y install python3-pip
pip install langflow
langflow --host 0.0.0.0 --port 7860
EOF
)

# Create a temporary file to store the startup script
tempfile=$(mktemp)
echo "$STARTUP_SCRIPT" > $tempfile

# Create the VM instance with the specified configuration and startup script
gcloud compute instances create $VM_NAME \
  --image-family $IMAGE_FAMILY \
  --image-project $IMAGE_PROJECT \
  --boot-disk-size $BOOT_DISK_SIZE \
  --machine-type=n1-standard-4 \
  --metadata-from-file startup-script=$tempfile \
  --zone $ZONE \
  --network $VPC_NAME \
  --subnet $SUBNET_NAME \
  --preemptible

# Remove the temporary file after the VM is created
rm $tempfile

By following these steps and ensuring your environment is correctly configured, you should be able to deploy your Langflow app correctly and resolve the issue of no response when running the Flow using cURL or the Python API. If the "unknown host" error persists, ensure that the server is running and accessible at the specified address (http://127.0.0.1:7864) [1][2][3][4][5].

To continue talking to Dosu, mention @dosu.

commented

The Issue is solved!