Sequentially (serially) execute celery task, UNIQUE BY (args, task_id) - python-3.x

Question

I want to execute celery tasks sequentially (serially, one by one) UNIQUE BY (taskid, args).
for example, we have Dockerfile and docker-compose.yml like these:
FROM python:3.8
WORKDIR /usr/src/app
ENV PYTHONDONTWRITEBYTECODE 1
ENV PYTHONUNBUFFERED 1
RUN pip install celery Redis flower
COPY . .
version: '3.3'
services:
handler:
build:
context: .
network: host
ports:
- 5000:5000
command: python main.py
volumes:
- .:/usr/src/app
environment:
- CELERY_BROKER_URL=redis://redis:6379/0
- CELERY_RESULT_BACKEND=redis://redis:6379/0
depends_on:
- redis
worker:
build:
context: .
network: host
command: celery --app main.app worker --uid=nobody --gid=nogroup --concurrency=1
volumes:
- .:/usr/src/app
environment:
- CELERY_BROKER_URL=redis://redis:6379/0
- CELERY_RESULT_BACKEND=redis://redis:6379/0
depends_on:
- handler
- redis
redis:
image: redis:6-alpine
and we have a python code like this:
import os
import time
from celery import Celery
app = Celery(__name__)
app.conf.broker_url = os.environ.get(
"CELERY_BROKER_URL", "redis://localhost:6379"
)
app.conf.result_backend = os.environ.get(
"CELERY_RESULT_BACKEND", "redis://localhost:6379"
)
#app.task()
def add(x):
time.sleep(10)
return x + 10
ret = add.apply_async(args=(1), task_id="foo") # no.1
ret = add.apply_async(args=(1), task_id="foo") # no.2 it should be queued until no.1 finished
ret = add.apply_async(args=(2), task_id="foo") # no.3 it should be executed asap because args is different from above
ret = add.apply_async(args=(2), task_id="bar") # no.4 it should be executed asap because task_id is different from each of them
here is a task that return 10 + args(x) after sleeps 10sec.
I want to execute tasks sequentially UNIQUE BY the pair of (task_id, args).
is it possible? any opinions are welcomed.
and also, I don't mind whether I use/don't use celery, so if there is an alternative way instead of celery, I'd love to know the idea.