Django-RQ Deployment Guide

Complete guide for deploying Django-RQ to production environments using Docker, Docker Compose, Kubernetes, or cloud platforms.

Architecture Overview

Django-RQ requires three main components for production:

Key Components

Django API: Web server (1+ processes)
RQ Workers: Background job processors (1+ per queue)
RQ Scheduler: Cron-like scheduler (1 process)
Redis: Message broker and result backend

Docker Compose Deployment

Production Setup

Complete docker-compose configuration for production:

docker-compose-production.yaml


version: '3.8'
 
services:
  # Django API Server
  django:
    image: your-registry/django-app:latest
    container_name: django-api
    restart: unless-stopped
    env_file: .env.prod
    environment:
      DJANGO_SETTINGS_MODULE: api.settings
    command: gunicorn api.wsgi:application --bind 0.0.0.0:8000 --workers 4
    ports:
      - "8000:8000"
    depends_on:
      - redis
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8000/cfg/health/"]
      interval: 30s
      timeout: 10s
      retries: 3
 
  # RQ Worker - High Priority
  rq-worker-high:
    image: your-registry/django-app:latest
    container_name: rq-worker-high
    restart: unless-stopped
    env_file: .env.prod
    environment:
      DJANGO_SETTINGS_MODULE: api.settings
    command: python manage.py rqworker high default
    depends_on:
      - redis
      - django
    deploy:
      replicas: 2  # Scale as needed
      resources:
        limits:
          cpus: '1.0'
          memory: 1G
    healthcheck:
      test: ["CMD-SHELL", "pgrep -f rqworker > /dev/null || exit 1"]
      interval: 30s
 
  # RQ Worker - Low Priority
  rq-worker-low:
    image: your-registry/django-app:latest
    container_name: rq-worker-low
    restart: unless-stopped
    env_file: .env.prod
    environment:
      DJANGO_SETTINGS_MODULE: api.settings
    command: python manage.py rqworker low knowledge
    depends_on:
      - redis
      - django
    deploy:
      resources:
        limits:
          cpus: '0.5'
          memory: 512M
 
  # RQ Scheduler
  rq-scheduler:
    image: your-registry/django-app:latest
    container_name: rq-scheduler
    restart: unless-stopped
    env_file: .env.prod
    environment:
      DJANGO_SETTINGS_MODULE: api.settings
    command: python manage.py rqscheduler
    depends_on:
      - redis
      - django
    deploy:
      resources:
        limits:
          cpus: '0.5'
          memory: 512M
 
  # Redis
  redis:
    image: redis:7-alpine
    container_name: redis
    restart: unless-stopped
    command: >
      redis-server
      --maxmemory 512mb
      --maxmemory-policy allkeys-lru
      --appendonly yes
    volumes:
      - redis_data:/data
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 10s
 
volumes:
  redis_data:

Start Production Services

All Services


# Start all services
docker compose -f docker-compose-production.yaml up -d
 
# Check status
docker compose -f docker-compose-production.yaml ps

Scale Workers


# Scale workers
docker compose -f docker-compose-production.yaml up -d --scale rq-worker-high=4
 
# Verify
docker ps | grep rq-worker

View Logs


# Worker logs
docker logs rq-worker-high -f
 
# Scheduler logs
docker logs rq-scheduler -f
 
# All logs
docker compose -f docker-compose-production.yaml logs -f

Local Development

Option 1: Docker Services

Use separate services for development:

docker-compose-local-services.yml


services:
  redis:
    image: redis:7-alpine
    ports:
      - "6379:6379"
    command: redis-server --maxmemory 256mb
 
  rq-worker:
    build: .
    command: python manage.py rqworker default high low knowledge
    env_file: .env.local
    depends_on:
      - redis
    volumes:
      - ./:/app
 
  rq-scheduler:
    build: .
    command: python manage.py rqscheduler
    env_file: .env.local
    depends_on:
      - redis
    volumes:
      - ./:/app


# Start services
docker compose -f docker-compose-local-services.yml up -d
 
# Run Django locally
python manage.py runserver

Option 2: Python Processes

Run everything locally for active development:

Manual


# Terminal 1: Django server
python manage.py runserver
 
# Terminal 2: RQ Worker
python manage.py rqworker default high low knowledge
 
# Terminal 3: RQ Scheduler (optional)
python manage.py rqscheduler

Using Makefile


# If django-cfg includes Makefile
 
# Start services
make rq
 
# View worker logs
make rq-worker
 
# View scheduler logs
make rq-scheduler
 
# Check queue stats
make rq-stats

Using Honcho/Foreman

Create Procfile:

Procfile.dev


web: python manage.py runserver 0.0.0.0:8000
worker: python manage.py rqworker default high low knowledge
scheduler: python manage.py rqscheduler


# Start all processes
honcho start -f Procfile.dev

Hot Reload in Development Workers automatically reload on code changes when using --with-scheduler flag:


python manage.py rqworker default --with-scheduler

Kubernetes Deployment

Complete K8s Manifests

Deployments

k8s/django-rq.yaml


apiVersion: apps/v1
kind: Deployment
metadata:
  name: django-api
spec:
  replicas: 3
  selector:
    matchLabels:
      app: django-api
  template:
    metadata:
      labels:
        app: django-api
    spec:
      containers:
      - name: django
        image: your-registry/django-app:latest
        command: ["gunicorn"]
        args: ["api.wsgi:application", "--bind", "0.0.0.0:8000", "--workers", "4"]
        ports:
        - containerPort: 8000
        envFrom:
        - configMapRef:
            name: django-config
        - secretRef:
            name: django-secrets
        resources:
          requests:
            memory: "512Mi"
            cpu: "500m"
          limits:
            memory: "1Gi"
            cpu: "1000m"
        livenessProbe:
          httpGet:
            path: /cfg/health/
            port: 8000
          initialDelaySeconds: 30
          periodSeconds: 10
 
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: rq-worker
spec:
  replicas: 4
  selector:
    matchLabels:
      app: rq-worker
  template:
    metadata:
      labels:
        app: rq-worker
    spec:
      containers:
      - name: worker
        image: your-registry/django-app:latest
        command: ["python", "manage.py", "rqworker"]
        args: ["default", "high", "low", "knowledge"]
        envFrom:
        - configMapRef:
            name: django-config
        - secretRef:
            name: django-secrets
        resources:
          requests:
            memory: "256Mi"
            cpu: "250m"
          limits:
            memory: "1Gi"
            cpu: "1000m"
 
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: rq-scheduler
spec:
  replicas: 1  # Only 1 scheduler needed
  selector:
    matchLabels:
      app: rq-scheduler
  template:
    metadata:
      labels:
        app: rq-scheduler
    spec:
      containers:
      - name: scheduler
        image: your-registry/django-app:latest
        command: ["python", "manage.py", "rqscheduler"]
        envFrom:
        - configMapRef:
            name: django-config
        - secretRef:
            name: django-secrets
        resources:
          requests:
            memory: "128Mi"
            cpu: "100m"
          limits:
            memory: "512Mi"
            cpu: "500m"

Services

k8s/services.yaml


apiVersion: v1
kind: Service
metadata:
  name: django-api
spec:
  selector:
    app: django-api
  ports:
  - port: 8000
    targetPort: 8000
  type: LoadBalancer
 
---
apiVersion: v1
kind: Service
metadata:
  name: redis
spec:
  selector:
    app: redis
  ports:
  - port: 6379
    targetPort: 6379
  type: ClusterIP

ConfigMap

k8s/configmap.yaml


apiVersion: v1
kind: ConfigMap
metadata:
  name: django-config
data:
  DJANGO_SETTINGS_MODULE: "api.settings"
  REDIS_URL: "redis://redis:6379/0"

Deploy to Kubernetes


# Apply all manifests
kubectl apply -f k8s/
 
# Check status
kubectl get pods
kubectl get svc
 
# View logs
kubectl logs -f deployment/rq-worker
 
# Scale workers
kubectl scale deployment/rq-worker --replicas=8

Scheduler Replicas Always keep rq-scheduler replicas at 1 to avoid duplicate scheduled jobs.

Cloud Platforms

AWS (ECS/Fargate)

ECS Task Definition

ecs-task-definition.json


{
  "family": "django-rq",
  "networkMode": "awsvpc",
  "requiresCompatibilities": ["FARGATE"],
  "cpu": "1024",
  "memory": "2048",
  "containerDefinitions": [
    {
      "name": "django-api",
      "image": "your-ecr-repo/django-app:latest",
      "command": ["gunicorn", "api.wsgi:application"],
      "portMappings": [{
        "containerPort": 8000,
        "protocol": "tcp"
      }],
      "environment": [
        {"name": "DJANGO_SETTINGS_MODULE", "value": "api.settings"},
        {"name": "REDIS_URL", "value": "redis://redis.cache.amazonaws.com:6379/0"}
      ],
      "logConfiguration": {
        "logDriver": "awslogs",
        "options": {
          "awslogs-group": "/ecs/django-rq",
          "awslogs-region": "us-east-1",
          "awslogs-stream-prefix": "django"
        }
      }
    },
    {
      "name": "rq-worker",
      "image": "your-ecr-repo/django-app:latest",
      "command": ["python", "manage.py", "rqworker", "default", "high", "low"],
      "environment": [
        {"name": "DJANGO_SETTINGS_MODULE", "value": "api.settings"},
        {"name": "REDIS_URL", "value": "redis://redis.cache.amazonaws.com:6379/0"}
      ]
    }
  ]
}

ElastiCache Setup


# Create Redis cluster
aws elasticache create-cache-cluster \
  --cache-cluster-id django-rq-redis \
  --cache-node-type cache.t3.medium \
  --engine redis \
  --num-cache-nodes 1 \
  --security-group-ids sg-xxxxx
 
# Get endpoint
aws elasticache describe-cache-clusters \
  --cache-cluster-id django-rq-redis \
  --show-cache-node-info

Google Cloud (Cloud Run + Cloud Tasks)

cloudrun-worker.yaml


apiVersion: serving.knative.dev/v1
kind: Service
metadata:
  name: django-rq-worker
spec:
  template:
    metadata:
      annotations:
        autoscaling.knative.dev/minScale: "1"
        autoscaling.knative.dev/maxScale: "10"
    spec:
      containers:
      - image: gcr.io/your-project/django-app:latest
        command: ["python", "manage.py", "rqworker"]
        args: ["default", "high", "low"]
        env:
        - name: REDIS_URL
          value: "redis://10.0.0.3:6379/0"  # MemoryStore IP
        resources:
          limits:
            memory: 1Gi
            cpu: "1"

Heroku

Procfile


web: gunicorn api.wsgi:application --bind 0.0.0.0:$PORT
worker: python manage.py rqworker default high low knowledge
scheduler: python manage.py rqscheduler


# Scale dynos
heroku ps:scale web=2 worker=4 scheduler=1
 
# Add Redis addon
heroku addons:create heroku-redis:premium-0

Configuration Examples

Environment Variables

Production (.env.prod)

.env.prod


# Django Settings
DJANGO_SETTINGS_MODULE=api.settings
SECRET_KEY=your-secret-key-min-50-chars
DEBUG=False
 
# Redis Configuration
REDIS_URL=redis://redis:6379/0
 
# Database
DATABASE_URL=postgresql://user:pass@postgres:5432/dbname
 
# Security
ALLOWED_HOSTS=api.example.com,*.example.com
CSRF_TRUSTED_ORIGINS=https://api.example.com
 
# RQ Settings (optional - auto-configured by django-cfg)
# RQ_SHOW_ADMIN_LINK=True
# RQ_PROMETHEUS_ENABLED=True

Local (.env.local)

.env.local


# Django Settings
DJANGO_SETTINGS_MODULE=api.settings
SECRET_KEY=local-dev-secret-key
DEBUG=True
 
# Redis Configuration
REDIS_URL=redis://localhost:6379/0
 
# Database
DATABASE_URL=postgresql://postgres:postgres@localhost:5432/django_dev
 
# Development
ALLOWED_HOSTS=*

Django-CFG Configuration

api/config.py


from django_cfg import DjangoConfig, DjangoRQConfig, RQQueueConfig, RQScheduleConfig
 
class ProductionConfig(DjangoConfig):
    debug: bool = False
    redis_url: str = "redis://redis:6379/0"
 
    django_rq: DjangoRQConfig = DjangoRQConfig(
        enabled=True,
        queues=[
            RQQueueConfig(queue="default", default_timeout=360),
            RQQueueConfig(queue="high", default_timeout=180),
            RQQueueConfig(queue="low", default_timeout=600),
            RQQueueConfig(queue="knowledge", default_timeout=1800),
        ],
        schedules=[
            RQScheduleConfig(
                func="apps.crypto.tasks.update_coin_prices",
                interval=300,  # Every 5 minutes
                queue="default",
            ),
        ],
        show_admin_link=True,
        prometheus_enabled=True,
    )

Monitoring in Production

Health Checks

Django API


# HTTP health check
curl http://localhost:8000/cfg/health/
 
# Expected response
{"status": "healthy", "checks": {"database": "ok", "redis": "ok"}}

RQ Worker


# Process check
pgrep -f rqworker > /dev/null && echo "Running" || echo "Down"
 
# Docker health check
docker exec rq-worker pgrep -f rqworker

Redis


# Ping test
redis-cli ping
# PONG
 
# Check memory
redis-cli info memory | grep used_memory_human

Logging

settings.py


LOGGING = {
    'version': 1,
    'handlers': {
        'console': {
            'class': 'logging.StreamHandler',
            'formatter': 'verbose',
        },
    },
    'formatters': {
        'verbose': {
            'format': '{levelname} {asctime} {module} {message}',
            'style': '{',
        },
    },
    'loggers': {
        'rq.worker': {
            'handlers': ['console'],
            'level': 'INFO',
        },
        'rq.scheduler': {
            'handlers': ['console'],
            'level': 'INFO',
        },
    },
}

Metrics Collection

Access Prometheus metrics at /django-rq/metrics/:


# Scrape metrics
curl http://localhost:8000/django-rq/metrics/
 
# Example metrics
rq_jobs_total{queue="default",status="finished"} 12450
rq_queue_length{queue="default"} 42
rq_workers_count{queue="default"} 4

Scaling Strategies

Horizontal Scaling

Docker Compose


# Scale workers
docker compose up -d --scale rq-worker=8
 
# Verify
docker ps | grep rq-worker

Kubernetes


# Scale deployment
kubectl scale deployment/rq-worker --replicas=10
 
# Auto-scaling (HPA)
kubectl autoscale deployment/rq-worker \
  --min=2 --max=20 \
  --cpu-percent=70

Manual


# Start multiple workers on different servers
# Server 1
python manage.py rqworker default high &
python manage.py rqworker default low &
 
# Server 2
python manage.py rqworker default high &
python manage.py rqworker default low &

Vertical Scaling


# Increase resources per worker
deploy:
  resources:
    limits:
      cpus: '2.0'      # Double CPU
      memory: 2G       # Double memory

Queue-Based Scaling


# Different workers for different queues
# High-priority workers (more resources)
python manage.py rqworker high --num-workers 4
 
# Low-priority workers (fewer resources)
python manage.py rqworker low knowledge --num-workers 2

Best Practices

Security

Production Security Checklist

✅ Use strong Redis password (requirepass in redis.conf)
✅ Restrict Redis network access (firewall rules)
✅ Enable TLS for Redis connections
✅ Use secrets management (not .env files)
✅ Limit worker permissions (non-root user)
✅ Enable authentication on RQ dashboard
✅ Use read-only Redis for monitoring

Performance

Performance Optimization

Worker Count: 1 worker per CPU core recommended
Queue Separation: Separate queues for different priorities
Result TTL: Set short TTL (300s) for frequent jobs
Connection Pooling: Use Redis connection pool (auto-enabled)
Job Timeout: Set appropriate timeouts per queue
Batch Processing: Group similar tasks together

Reliability

High Availability

Redis: Use Redis Sentinel or Cluster for HA
Workers: Run at least 2 workers per queue
Scheduler: Run 1 scheduler with restart policy
Monitoring: Set up alerts for queue depth
Health Checks: Monitor worker processes

Troubleshooting

Workers Not Processing Jobs

Check Redis


# Test Redis connection
python manage.py shell
>>> import django_rq
>>> conn = django_rq.get_connection()
>>> conn.ping()
True

Check Queue


# View queue statistics
python manage.py rqstats
 
# Check queue contents
python manage.py shell
>>> from django_rq import get_queue
>>> queue = get_queue('default')
>>> print(f"Jobs in queue: {len(queue)}")

Check Logs


# Docker logs
docker logs rq-worker --tail 100 -f
 
# Process logs (if using systemd)
journalctl -u rq-worker -f

Common Issues

Issue	Cause	Solution
Jobs stuck in queue	No workers running	Start workers: `python manage.py rqworker`
Jobs failing silently	Exceptions not logged	Check failed registry in admin
High memory usage	Result TTL too long	Reduce `default_result_ttl` in config
Slow job processing	Too few workers	Scale workers horizontally
Scheduler not running	Process crashed	Check logs, restart scheduler

Migration Checklist

Migrating from ReArq or Celery? Use this checklist:

Migration from ReArq

ReArq → Django-RQ Migration

Update docker-compose (remove rearq services)
Update config.py (DjangoRQConfig)
Convert async tasks to sync (if needed)
Update enqueue calls (django_rq.enqueue())
Start RQ workers (python manage.py rqworker)
Test scheduled jobs
Update monitoring dashboards
Remove ReArq dependencies

Migration from Celery

Celery → Django-RQ Migration

Update config (remove Celery config)
Convert @task to plain functions
Replace task.delay() with queue.enqueue()
Replace periodic tasks with RQ schedules
Update monitoring (Flower → RQ Admin)
Test job execution
Migrate cron schedules
Remove Celery dependencies

Next Steps

Production Ready? Review monitoring guide
Need Examples? Check examples guide
Advanced Config? See configuration reference
Architecture? Read architecture guide

Django-RQ Deployment Guide

Quick Navigation

Docker Compose

Kubernetes

Cloud Platforms

Local Development

Architecture Overview

Docker Compose Deployment

Production Setup

Start Production Services

All Services

Scale Workers

View Logs

Local Development

Option 1: Docker Services

Option 2: Python Processes

Manual

Using Makefile

Using Honcho/Foreman

Kubernetes Deployment

Complete K8s Manifests

Deployments

Services

ConfigMap

Deploy to Kubernetes

Cloud Platforms

AWS (ECS/Fargate)

ECS Task Definition

ElastiCache Setup

Google Cloud (Cloud Run + Cloud Tasks)

Heroku

Configuration Examples

Environment Variables

Production (.env.prod)

Local (.env.local)

Django-CFG Configuration

Monitoring in Production

Health Checks

Django API

RQ Worker

Redis

Logging

Metrics Collection

Scaling Strategies

Horizontal Scaling

Docker Compose

Kubernetes

Manual

Vertical Scaling

Queue-Based Scaling

Best Practices

Security

Performance

Reliability

Troubleshooting

Workers Not Processing Jobs

Check Redis

Check Queue

Check Logs

Common Issues

Migration Checklist

Migration from ReArq

ReArq → Django-RQ Migration

Migration from Celery

Celery → Django-RQ Migration

Next Steps