Production Deployment

Deploy DeerFlow to production with enterprise-grade security, reliability, and performance.

Architecture Overview

┌────────────────────────────────────────────────────────────┐
│                    Load Balancer / CDN                      │
│                   (Cloudflare, AWS ALB)                     │
└──────────────────────────┬─────────────────────────────────┘
                           │
                           ▼
┌────────────────────────────────────────────────────────────┐
│                    Nginx / API Gateway                      │
│              (SSL termination, rate limiting)               │
└───────┬──────────────────┬─────────────────┬───────────────┘
        │                  │                 │
        ▼                  ▼                 ▼
┌──────────────┐  ┌──────────────┐  ┌──────────────┐
│  Frontend    │  │   Gateway    │  │  LangGraph   │
│  (Replicas)  │  │  (Replicas)  │  │  (Replicas)  │
└──────────────┘  └──────┬───────┘  └──────┬───────┘
                         │                  │
                         ▼                  ▼
              ┌──────────────────────────────────┐
              │      Shared Storage (NFS/EFS)    │
              │  - Thread data                   │
              │  - Skills                        │
              │  - Artifacts                     │
              └──────────────────────────────────┘
                         │
                         ▼
              ┌──────────────────────────────────┐
              │     Sandbox Provisioner          │
              │   (Kubernetes-based sandboxes)   │
              └──────────────────────────────────┘

Pre-deployment Checklist

Infrastructure

Kubernetes cluster (EKS, GKE, AKS, or self-hosted)
Container registry (ECR, GCR, Docker Hub)
Load balancer (ALB, GCE LB, Nginx Ingress)
TLS certificates (Let’s Encrypt, AWS ACM)
Shared storage (NFS, EFS, Cloud Filestore)
Database for checkpoints (PostgreSQL, Redis)
Monitoring stack (Prometheus, Grafana)
Logging aggregation (ELK, CloudWatch, Loki)
Backup solution (Velero, cloud snapshots)

Security

API keys stored in secrets manager (Vault, AWS Secrets Manager)
Network policies configured
RBAC policies defined
Security scanning enabled (Snyk, Trivy)
SSL/TLS certificates deployed
Rate limiting configured
DDoS protection enabled
Audit logging enabled

Configuration

config.yaml prepared for production
Environment-specific settings configured
Resource limits defined
Auto-scaling policies configured
Health check endpoints tested
Backup retention policies defined

Production Configuration

config.yaml

Production-ready configuration:

models:
  - name: gpt-4
    display_name: GPT-4
    use: langchain_openai:ChatOpenAI
    model: gpt-4
    api_key: $OPENAI_API_KEY  # From secrets manager
    max_tokens: 4096
    temperature: 0.7
    timeout: 120  # Increased timeout
    max_retries: 3  # Retry failed requests

tool_groups:
  - name: web
  - name: file:read
  - name: file:write
  - name: bash

tools:
  - name: web_search
    group: web
    use: src.community.tavily.tools:web_search_tool
    max_results: 5
    timeout: 30
  
  - name: web_fetch
    group: web
    use: src.community.jina_ai.tools:web_fetch_tool
    timeout: 30

# Use Kubernetes-based sandbox with provisioner
sandbox:
  use: src.community.aio_sandbox:AioSandboxProvider
  provisioner_url: http://provisioner:8002
  # Resource limits enforced by provisioner pod config

subagents:
  timeout_seconds: 1800  # 30 minutes for complex tasks
  agents:
    general-purpose:
      timeout_seconds: 3600  # 1 hour for research
    bash:
      timeout_seconds: 600  # 10 minutes for commands

skills:
  container_path: /mnt/skills

title:
  enabled: true
  max_words: 6
  max_chars: 60
  model_name: null

# Aggressive summarization for long conversations
summarization:
  enabled: true
  model_name: null
  trigger:
    - type: tokens
      value: 50000  # Trigger earlier in production
    - type: fraction
      value: 0.7  # When 70% of context used
  keep:
    type: messages
    value: 20  # Keep more recent messages
  trim_tokens_to_summarize: 50000

# Memory persistence
memory:
  enabled: true
  storage_path: /mnt/shared/memory.json
  debounce_seconds: 60
  model_name: null
  max_facts: 200
  fact_confidence_threshold: 0.8  # Higher threshold
  injection_enabled: true
  max_injection_tokens: 3000

Environment Variables

Store sensitive data in secrets:

# API Keys (from secrets manager)
OPENAI_API_KEY=secret-ref:openai-key
ANTHROPIC_API_KEY=secret-ref:anthropic-key
TAVILY_API_KEY=secret-ref:tavily-key

# Application settings
NODE_ENV=production
CI=true

# Logging
LOG_LEVEL=INFO
LOG_FORMAT=json

# Monitoring
ENABLE_METRICS=true
METRICS_PORT=9090

# Tracing (optional)
OTEL_EXPORTER_OTLP_ENDPOINT=http://jaeger:4317

Container Images

Build Production Images

Frontend Production Dockerfile

Create frontend/Dockerfile.prod:

# Build stage
FROM node:22-alpine AS builder

WORKDIR /app

# Install pnpm
RUN corepack enable && corepack install -g pnpm@10.26.2

# Copy package files
COPY frontend/package.json frontend/pnpm-lock.yaml ./

# Install dependencies
RUN pnpm install --frozen-lockfile --prod=false

# Copy source
COPY frontend/ .

# Build
RUN pnpm run build

# Production stage
FROM node:22-alpine

WORKDIR /app

# Install pnpm
RUN corepack enable && corepack install -g pnpm@10.26.2

# Copy built assets and dependencies
COPY --from=builder /app/.next ./.next
COPY --from=builder /app/public ./public
COPY --from=builder /app/package.json ./package.json
COPY --from=builder /app/node_modules ./node_modules

# Create non-root user
RUN addgroup -g 1001 -S nodejs && \
    adduser -S nextjs -u 1001 && \
    chown -R nextjs:nodejs /app

USER nextjs

EXPOSE 3000

ENV NODE_ENV=production
ENV PORT=3000

CMD ["pnpm", "start"]

Backend Production Dockerfile

Create backend/Dockerfile.prod:

FROM python:3.12-slim

# Install system dependencies
RUN apt-get update && apt-get install -y --no-install-recommends \
    curl \
    build-essential \
    && rm -rf /var/lib/apt/lists/*

# Install uv
RUN curl -LsSf https://astral.sh/uv/install.sh | sh
ENV PATH="/root/.local/bin:$PATH"

WORKDIR /app

# Copy backend
COPY backend ./backend

# Install production dependencies only
RUN --mount=type=cache,target=/root/.cache/uv \
    sh -c "cd backend && uv sync --no-dev"

# Create non-root user
RUN useradd -m -u 1001 deerflow && \
    chown -R deerflow:deerflow /app

USER deerflow

EXPOSE 8001 2024

ENV PYTHONUNBUFFERED=1

# Multi-worker uvicorn for production
CMD ["sh", "-c", "cd backend && uv run uvicorn src.gateway.app:app --host 0.0.0.0 --port 8001 --workers 4 --log-level info"]

Build and Push

# Build images
docker build -t your-registry/deer-flow-frontend:v2.0 -f frontend/Dockerfile.prod .
docker build -t your-registry/deer-flow-backend:v2.0 -f backend/Dockerfile.prod .

# Push to registry
docker push your-registry/deer-flow-frontend:v2.0
docker push your-registry/deer-flow-backend:v2.0

# Tag as latest
docker tag your-registry/deer-flow-frontend:v2.0 your-registry/deer-flow-frontend:latest
docker tag your-registry/deer-flow-backend:v2.0 your-registry/deer-flow-backend:latest
docker push your-registry/deer-flow-frontend:latest
docker push your-registry/deer-flow-backend:latest

Kubernetes Deployment

Namespace

apiVersion: v1
kind: Namespace
metadata:
  name: deer-flow-prod
  labels:
    name: deer-flow-prod
    environment: production

Frontend Deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: frontend
  namespace: deer-flow-prod
spec:
  replicas: 3
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxSurge: 1
      maxUnavailable: 0
  selector:
    matchLabels:
      app: deer-flow-frontend
  template:
    metadata:
      labels:
        app: deer-flow-frontend
    spec:
      containers:
      - name: frontend
        image: your-registry/deer-flow-frontend:v2.0
        ports:
        - containerPort: 3000
        env:
        - name: NODE_ENV
          value: "production"
        - name: NEXT_PUBLIC_API_URL
          value: "https://api.deerflow.example.com"
        resources:
          requests:
            cpu: 500m
            memory: 512Mi
          limits:
            cpu: 1000m
            memory: 1Gi
        livenessProbe:
          httpGet:
            path: /
            port: 3000
          initialDelaySeconds: 30
          periodSeconds: 10
        readinessProbe:
          httpGet:
            path: /
            port: 3000
          initialDelaySeconds: 5
          periodSeconds: 5
      affinity:
        podAntiAffinity:
          preferredDuringSchedulingIgnoredDuringExecution:
          - weight: 100
            podAffinityTerm:
              labelSelector:
                matchExpressions:
                - key: app
                  operator: In
                  values:
                  - deer-flow-frontend
              topologyKey: kubernetes.io/hostname

Gateway Deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: gateway
  namespace: deer-flow-prod
spec:
  replicas: 3
  selector:
    matchLabels:
      app: deer-flow-gateway
  template:
    metadata:
      labels:
        app: deer-flow-gateway
    spec:
      containers:
      - name: gateway
        image: your-registry/deer-flow-backend:v2.0
        command: ["sh", "-c"]
        args:
        - |
          cd backend && 
          uv run uvicorn src.gateway.app:app \
            --host 0.0.0.0 --port 8001 \
            --workers 4 \
            --log-level info
        ports:
        - containerPort: 8001
        env:
        - name: OPENAI_API_KEY
          valueFrom:
            secretKeyRef:
              name: deer-flow-secrets
              key: openai-api-key
        volumeMounts:
        - name: config
          mountPath: /app/config.yaml
          subPath: config.yaml
        - name: shared-storage
          mountPath: /mnt/shared
        resources:
          requests:
            cpu: 1000m
            memory: 1Gi
          limits:
            cpu: 2000m
            memory: 2Gi
        livenessProbe:
          httpGet:
            path: /health
            port: 8001
          initialDelaySeconds: 30
          periodSeconds: 10
        readinessProbe:
          httpGet:
            path: /health
            port: 8001
          initialDelaySeconds: 10
          periodSeconds: 5
      volumes:
      - name: config
        configMap:
          name: deer-flow-config
      - name: shared-storage
        persistentVolumeClaim:
          claimName: deer-flow-shared-pvc

LangGraph Deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: langgraph
  namespace: deer-flow-prod
spec:
  replicas: 3
  selector:
    matchLabels:
      app: deer-flow-langgraph
  template:
    metadata:
      labels:
        app: deer-flow-langgraph
    spec:
      containers:
      - name: langgraph
        image: your-registry/deer-flow-backend:v2.0
        command: ["sh", "-c"]
        args:
        - |
          cd backend && 
          uv run langgraph start \
            --host 0.0.0.0 --port 2024
        ports:
        - containerPort: 2024
        env:
        - name: OPENAI_API_KEY
          valueFrom:
            secretKeyRef:
              name: deer-flow-secrets
              key: openai-api-key
        volumeMounts:
        - name: config
          mountPath: /app/config.yaml
          subPath: config.yaml
        - name: shared-storage
          mountPath: /mnt/shared
        - name: threads
          mountPath: /app/backend/.deer-flow
        resources:
          requests:
            cpu: 2000m
            memory: 2Gi
          limits:
            cpu: 4000m
            memory: 4Gi
        livenessProbe:
          httpGet:
            path: /
            port: 2024
          initialDelaySeconds: 30
          periodSeconds: 10
      volumes:
      - name: config
        configMap:
          name: deer-flow-config
      - name: shared-storage
        persistentVolumeClaim:
          claimName: deer-flow-shared-pvc
      - name: threads
        persistentVolumeClaim:
          claimName: deer-flow-threads-pvc

Services

---
apiVersion: v1
kind: Service
metadata:
  name: frontend
  namespace: deer-flow-prod
spec:
  type: ClusterIP
  ports:
  - port: 3000
    targetPort: 3000
  selector:
    app: deer-flow-frontend
---
apiVersion: v1
kind: Service
metadata:
  name: gateway
  namespace: deer-flow-prod
spec:
  type: ClusterIP
  ports:
  - port: 8001
    targetPort: 8001
  selector:
    app: deer-flow-gateway
---
apiVersion: v1
kind: Service
metadata:
  name: langgraph
  namespace: deer-flow-prod
spec:
  type: ClusterIP
  ports:
  - port: 2024
    targetPort: 2024
  selector:
    app: deer-flow-langgraph

Ingress

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: deer-flow-ingress
  namespace: deer-flow-prod
  annotations:
    kubernetes.io/ingress.class: nginx
    cert-manager.io/cluster-issuer: letsencrypt-prod
    nginx.ingress.kubernetes.io/ssl-redirect: "true"
    nginx.ingress.kubernetes.io/proxy-body-size: 100m
    nginx.ingress.kubernetes.io/proxy-read-timeout: "600"
    nginx.ingress.kubernetes.io/proxy-send-timeout: "600"
spec:
  tls:
  - hosts:
    - deerflow.example.com
    - api.deerflow.example.com
    secretName: deer-flow-tls
  rules:
  - host: deerflow.example.com
    http:
      paths:
      - path: /
        pathType: Prefix
        backend:
          service:
            name: frontend
            port:
              number: 3000
  - host: api.deerflow.example.com
    http:
      paths:
      - path: /api/langgraph
        pathType: Prefix
        backend:
          service:
            name: langgraph
            port:
              number: 2024
      - path: /api
        pathType: Prefix
        backend:
          service:
            name: gateway
            port:
              number: 8001

Scaling

Horizontal Pod Autoscaler

---
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: frontend-hpa
  namespace: deer-flow-prod
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: frontend
  minReplicas: 3
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 70
  - type: Resource
    resource:
      name: memory
      target:
        type: Utilization
        averageUtilization: 80
---
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: gateway-hpa
  namespace: deer-flow-prod
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: gateway
  minReplicas: 3
  maxReplicas: 20
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 75
---
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: langgraph-hpa
  namespace: deer-flow-prod
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: langgraph
  minReplicas: 3
  maxReplicas: 20
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 80

Cluster Autoscaler

Enable cluster autoscaler for node scaling:

apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-autoscaler
  namespace: kube-system
data:
  min-nodes: "3"
  max-nodes: "20"
  scale-down-delay-after-add: "10m"
  scale-down-unneeded-time: "10m"

Monitoring

Prometheus ServiceMonitor

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: deer-flow-metrics
  namespace: deer-flow-prod
spec:
  selector:
    matchLabels:
      app: deer-flow-gateway
  endpoints:
  - port: metrics
    interval: 30s
    path: /metrics

Grafana Dashboard

Key metrics to monitor:

Request rate: Requests per second
Response time: P50, P95, P99 latencies
Error rate: 4xx/5xx responses
Active threads: Number of agent threads
Sandbox usage: Active sandbox pods
Resource utilization: CPU, memory, disk
Queue depth: Pending agent tasks

Alerts

apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  name: deer-flow-alerts
  namespace: deer-flow-prod
spec:
  groups:
  - name: deer-flow
    interval: 30s
    rules:
    - alert: HighErrorRate
      expr: rate(http_requests_total{status=~"5.."}[5m]) > 0.05
      for: 5m
      annotations:
        summary: High error rate detected
    
    - alert: HighLatency
      expr: histogram_quantile(0.95, rate(http_request_duration_seconds_bucket[5m])) > 5
      for: 5m
      annotations:
        summary: High latency detected (P95 > 5s)
    
    - alert: PodCrashLooping
      expr: rate(kube_pod_container_status_restarts_total[15m]) > 0
      for: 5m
      annotations:
        summary: Pod is crash looping

Security

Network Policies

apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
  name: deer-flow-network-policy
  namespace: deer-flow-prod
spec:
  podSelector:
    matchLabels:
      app: deer-flow-gateway
  policyTypes:
  - Ingress
  - Egress
  ingress:
  - from:
    - podSelector:
        matchLabels:
          app: deer-flow-frontend
    - podSelector:
        matchLabels:
          app: deer-flow-langgraph
    ports:
    - protocol: TCP
      port: 8001
  egress:
  - to:
    - podSelector:
        matchLabels:
          app: deer-flow-provisioner
    ports:
    - protocol: TCP
      port: 8002
  - to:
    - namespaceSelector: {}
    ports:
    - protocol: TCP
      port: 443  # HTTPS for external APIs

Pod Security Policy

apiVersion: policy/v1beta1
kind: PodSecurityPolicy
metadata:
  name: deer-flow-psp
spec:
  privileged: false
  allowPrivilegeEscalation: false
  requiredDropCapabilities:
  - ALL
  runAsUser:
    rule: MustRunAsNonRoot
  seLinux:
    rule: RunAsAny
  fsGroup:
    rule: RunAsAny
  volumes:
  - configMap
  - secret
  - persistentVolumeClaim
  - emptyDir

Secrets Management

Use external secrets operator:

apiVersion: external-secrets.io/v1beta1
kind: SecretStore
metadata:
  name: aws-secrets-manager
  namespace: deer-flow-prod
spec:
  provider:
    aws:
      service: SecretsManager
      region: us-east-1
---
apiVersion: external-secrets.io/v1beta1
kind: ExternalSecret
metadata:
  name: deer-flow-secrets
  namespace: deer-flow-prod
spec:
  refreshInterval: 1h
  secretStoreRef:
    name: aws-secrets-manager
    kind: SecretStore
  target:
    name: deer-flow-secrets
    creationPolicy: Owner
  data:
  - secretKey: openai-api-key
    remoteRef:
      key: prod/deerflow/openai-api-key
  - secretKey: anthropic-api-key
    remoteRef:
      key: prod/deerflow/anthropic-api-key

Backup and Disaster Recovery

Velero Backup

# Install Velero
velero install \
  --provider aws \
  --plugins velero/velero-plugin-for-aws:v1.9.0 \
  --bucket deer-flow-backups \
  --backup-location-config region=us-east-1 \
  --snapshot-location-config region=us-east-1

# Create backup schedule
velero schedule create deer-flow-daily \
  --schedule="0 2 * * *" \
  --include-namespaces deer-flow-prod \
  --ttl 720h0m0s

# Manual backup
velero backup create deer-flow-backup-$(date +%Y%m%d) \
  --include-namespaces deer-flow-prod

Thread Data Backup

#!/bin/bash
# backup-threads.sh

BACKUP_DIR="/backups/threads"
THREADS_DIR="/mnt/shared/threads"
RETENTION_DAYS=30

# Create timestamped backup
TIMESTAMP=$(date +%Y%m%d-%H%M%S)
tar -czf "${BACKUP_DIR}/threads-${TIMESTAMP}.tar.gz" "${THREADS_DIR}"

# Upload to S3
aws s3 cp "${BACKUP_DIR}/threads-${TIMESTAMP}.tar.gz" \
  s3://deer-flow-backups/threads/

# Clean old backups
find "${BACKUP_DIR}" -name "threads-*.tar.gz" -mtime +${RETENTION_DAYS} -delete

Restore Procedure

# Restore from Velero
velero restore create --from-backup deer-flow-backup-20260304

# Restore thread data
aws s3 cp s3://deer-flow-backups/threads/threads-20260304-020000.tar.gz .
tar -xzf threads-20260304-020000.tar.gz -C /mnt/shared/

Performance Optimization

Caching

Redis for session state:

env:
- name: REDIS_URL
  value: redis://redis:6379/0

CDN for static assets:
- Cloudflare
- AWS CloudFront
- Fastly

Response caching:

# In gateway app
from fastapi_cache import FastAPICache
from fastapi_cache.backends.redis import RedisBackend

@app.on_event("startup")
async def startup():
    redis = await aioredis.create_redis_pool("redis://redis:6379")
    FastAPICache.init(RedisBackend(redis), prefix="deer-flow-cache")

Database Optimization

For checkpoint storage:

# PostgreSQL for LangGraph checkpoints
apiVersion: v1
kind: Service
metadata:
  name: postgres
  namespace: deer-flow-prod
spec:
  ports:
  - port: 5432
  selector:
    app: postgres
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: postgres
  namespace: deer-flow-prod
spec:
  serviceName: postgres
  replicas: 1
  selector:
    matchLabels:
      app: postgres
  template:
    spec:
      containers:
      - name: postgres
        image: postgres:16-alpine
        env:
        - name: POSTGRES_DB
          value: deerflow
        - name: POSTGRES_USER
          valueFrom:
            secretKeyRef:
              name: postgres-secret
              key: username
        - name: POSTGRES_PASSWORD
          valueFrom:
            secretKeyRef:
              name: postgres-secret
              key: password
        volumeMounts:
        - name: postgres-storage
          mountPath: /var/lib/postgresql/data
        resources:
          requests:
            cpu: 1000m
            memory: 2Gi
          limits:
            cpu: 2000m
            memory: 4Gi
  volumeClaimTemplates:
  - metadata:
      name: postgres-storage
    spec:
      accessModes: ["ReadWriteOnce"]
      resources:
        requests:
          storage: 100Gi

Cost Optimization

Resource Right-sizing

Monitor and adjust resource requests/limits:

# Analyze resource usage
kubectl top pods -n deer-flow-prod

# View recommendations
kubectl describe vpa -n deer-flow-prod

Spot Instances

Use spot instances for sandbox pods:

spec:
  nodeSelector:
    node.kubernetes.io/instance-type: spot
  tolerations:
  - key: spot
    operator: Equal
    value: "true"
    effect: NoSchedule

Storage Tiering

# Hot data on SSD
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: deer-flow-hot-storage
spec:
  storageClassName: fast-ssd
  accessModes:
  - ReadWriteMany
  resources:
    requests:
      storage: 100Gi

# Cold data on HDD
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: deer-flow-cold-storage
spec:
  storageClassName: standard-hdd
  accessModes:
  - ReadWriteMany
  resources:
    requests:
      storage: 1Ti

Troubleshooting

Check Pod Status

kubectl get pods -n deer-flow-prod
kubectl describe pod <pod-name> -n deer-flow-prod
kubectl logs <pod-name> -n deer-flow-prod --tail=100 -f

Check Resource Usage

kubectl top nodes
kubectl top pods -n deer-flow-prod

Debug Network Issues

# Test service connectivity
kubectl run -it --rm debug --image=nicolaka/netshoot --restart=Never -- bash
curl http://gateway:8001/health

# Check network policies
kubectl get networkpolicies -n deer-flow-prod

Performance Profiling

# Enable profiling in gateway
kubectl exec -it gateway-xxx -n deer-flow-prod -- bash
cd backend
uv run python -m cProfile -o profile.stats src/gateway/app.py

​Architecture Overview

​Pre-deployment Checklist

​Infrastructure

​Security

​Configuration

​Production Configuration

​config.yaml

​Environment Variables

​Container Images

​Build Production Images

​Frontend Production Dockerfile

​Backend Production Dockerfile

​Build and Push

​Kubernetes Deployment

​Namespace

​Frontend Deployment

​Gateway Deployment

​LangGraph Deployment

​Services

​Ingress

​Scaling

​Horizontal Pod Autoscaler

​Cluster Autoscaler

​Monitoring

​Prometheus ServiceMonitor

​Grafana Dashboard

​Alerts

​Security

​Network Policies

​Pod Security Policy

​Secrets Management

​Backup and Disaster Recovery

​Velero Backup

​Thread Data Backup

​Restore Procedure

​Performance Optimization

​Caching

​Database Optimization

​Cost Optimization

​Resource Right-sizing

​Spot Instances

​Storage Tiering

​Troubleshooting

​Check Pod Status

​Check Resource Usage

​Debug Network Issues

​Performance Profiling

​Next Steps

Docker Deployment

Kubernetes Deployment

​See Also

Architecture Overview

Pre-deployment Checklist

Infrastructure

Security

Configuration

Production Configuration

config.yaml

Environment Variables

Container Images

Build Production Images

Frontend Production Dockerfile

Backend Production Dockerfile

Build and Push

Kubernetes Deployment

Namespace

Frontend Deployment

Gateway Deployment

LangGraph Deployment

Services

Ingress

Scaling

Horizontal Pod Autoscaler

Cluster Autoscaler

Monitoring

Prometheus ServiceMonitor

Grafana Dashboard

Alerts

Security

Network Policies

Pod Security Policy

Secrets Management

Backup and Disaster Recovery

Velero Backup

Thread Data Backup

Restore Procedure

Performance Optimization

Caching

Database Optimization

Cost Optimization

Resource Right-sizing

Spot Instances

Storage Tiering

Troubleshooting

Check Pod Status

Check Resource Usage

Debug Network Issues

Performance Profiling

Next Steps

See Also