Running Centralized Service on Elastic Beanstalk - django

Question

I have setup a Django server on AWS Elastic Beanstalk. In that server, I have following services
Application server
Celery Worker
Celery Beat
I have using Docker to deploy my application which means I build my docker image, us that image to run all three services. My Dockerrun.aws.json file is below.
{
"AWSEBDockerrunVersion": 2,
"containerDefinitions": [
{
"command": [
"sh",
"-c",
"./entry_point.sh && gunicorn Project.wsgi:application -w 2 -b :8000 --timeout 120 --graceful-timeout 120 --worker-class gevent"
],
"environment": [
],
"essential": true,
"image": "695189796512.dkr.ecr.us-west-2.amazonaws.com/Project-20181107174734",
"name": "app",
"memory": 500,
"portMappings": [
{
"containerPort": 8000,
"hostPort": 80
}
]
},
{
"command": [
"celery",
"-A",
"Project",
"beat",
"--loglevel=info",
"--uid",
"django"
],
"environment": [
],
"essential": true,
"image": "695189796512.dkr.ecr.us-west-2.amazonaws.com/Project-20181107174734",
"memory": 200,
"name": "celery-beat"
},
{
"command": [
"celery",
"-A",
"Project",
"worker",
"--loglevel=info",
"--uid",
"django"
],
"environment": [
],
"essential": true,
"image": "695189796512.dkr.ecr.us-west-2.amazonaws.com/Project-20181107174734",
"memory": 200,
"name": "celery-worker"
}
],
"family": "",
"volumes": []
}
Problem:
This configuration is working fine. But the problem with this configuration is that on all my nodes, all three services run. When load increases, my server scales up to multiple nodes and all services run on each node. I can have multiple celery worker running on server as my task queue is same (I am using SQS). But I only want one Celery Beat service running on all my nodes because multiple Beat services can cause task added to queue multiple time.
What I wanted to do is to run a centralized service in my server. I want help that how I can achieve this in my current setup.
Also I want to know is there any problem in using multiple instance of Celery Worker on my server?.